Home

DOCTORAL THE SIS - Tesis doctorales en red

1. 222222222e22e 32 Fig 13 Comparison of incremental displacements calculated with the PLAXIS model and experimentally obtained for the representative stages Of Wall W45 ooo eeceecesceseeeseeeeeeeeseeeaeceaecnseeseeeseeeseeeees 33 Fig 14 Horizontal displacements and bending moments obtained by the adjusted PLAXIS model for Wall WS Samitier Sa ta AN a 34 Fig 15 Horizontal displacements and bending moments obtained by the adjusted PLAXIS model for VVall WA Diced ets atri ara ia trat ts ec dte nt enca Es 34 Fig 16 Horizontal displacements obtained by the PLAXIS model for the T BLVV45 design 36 Fig 17 Maximum displacements for the three Comparative CASES ooooccnocccooccnononancnoncnononnnononoconccona ran cona cra 36 Fig 18 Bending moments obtained by the PLAXIS model for the T BLVV45 design example 37 Fig 19 Bending moment envelopes for the three Comparative CASES ooooccnccccocnnocononcnonononoconoconacnnncnnnrnn conan 38 Fig 20 a Sectional discretization b SFRC and c steel bar constitutive equations eeeo0eu2eo see 43 Fig 21 a Model geometry Anchorages and slabs positions b Simple Section c Compound Section 44 Fig 22 Finite element model mesh and main elements 00 0 cece eeceeeeeeeeeeeeeeeeeeeaeecaeceaeceseceeeeeeeeeeeeseneeenees 46 Fig 23 Wall displacements Anc Out stage for the three Walls eee eseesecsecsseceseceseceeee
2. ML o BLoo 10 BLs5410 EI SS 475 2 475 2 366 0 MN m sc 754 6 604 2 EA SS 15840 15 840 14 520 MN SC 25 153 23 967 Value calculated to keep the EI EA ratio unchanged The soil elements were 15 node triangular finite elements under plane strain A sandy soil was chosen for this analysis as water filtration problems are more likely in a permeable soil The soil parameters and the values used for the selected model correspond to the Lake sand layer taken from an experimental case reported in the literature Hashash et al 2010 The coefficient of lateral earth pressure at rest Ro was automatically estimated by the program using the expression of Jaky Terzaghi et al 1996 A total unit weight of y 20 0 kN m was selected for the soil Interface elements were set out for the soil in contact with the plate elements with a strength reduction factor for soil structure interface of Rinter 0 66 Khoiri amp Ou 2013 and continued 1 0 m below the bottom end of the walls as suggested by Luis Segura Castillo Structural and sectional analysis 47 PLAXIS PLAXIS 2D 2010b A Drained analysis was used even though the phreatic level was below the model boundaries and therefore no water flow was considered The permanent wall supports were modelled with fixed end anchors A normal stiffness of EA 7 26 10 kN m equivalent to a 22 cm thick massive slab with an equivalent support length of 10 m the length from
3. 16 mss ML 50 16 174 M3 BLoo 10 17 SS d MS BLss 10 ds CS 19 E Ma BLoo 10 M BLss 10 20 20 350 300 250 200 150 100 50 0 50 100 150 200 250 300 350 300 250 200 150 100 50 0 50 100 150 200 250 300 350 Bending Moments RN m Bending Moments RN m Fig 24 Bending moments a Envelopes for the three vvall types b Representative stages and envelopes for the BL60 10 wall type c envelope areas for all three wall types o 8 ML so BL 60110 BL 55 10 Luis Segura Castillo Structural and sectional analysis 49 Comparing the bi layer walls it can be seen that the BL o 0 shows larger moments than the BLss 10 along the whole length of the wall This is a consequence of the greater stiffness of the RC layer and therefore greater stiffness both in the SS and in the CS cross section The MLeo and BL o 10 wall types show practically identical envelopes in the embedded section of the wall between depths 12 5m and 20 0m In this section both walls have the same cross section i e the RC layer for all the stages The biggest differences between these wall types was registered in the centre of the walls between depths 5 0m and 12 0m in which the BLo 9 envelope was larger As the SFRC layer is sprayed the upper stretches become stiffer diminishing any relative collaboration of the embedded part of the wall The bending moments of the BL e019 wall t
4. 0 00 1 10 100 Age of 21d phase concrete days Shear strength t MPa Fig 48 Shear stress vs age of second phase concrete surface preparation with milling MP and saturated surface SP Bi layer diaphragm walls Experimental and numerical analysis 88 Chapter 5 In the case of MP it can be clearly observed that the shear stress value increases with age The linear regression line traced in Fig 48 shows an excellent fit with Plowman s modified function TS A B log maturity 8 where A and B are constants to be determined and maturity is defined by the Nurse Saul function maturity E T T9 At 9 where To is the datum temperature usually 10 C and At is a time interval Therefore in agreement with Delatte s study Delatte et al 2000 the evolution of shear stress for the second phase carried out with sprayed concrete fits in with the maturity model if it is assumed once again that the average temperature of concrete for the period of time under study is constant Homogeneous low variations of bond strength were obtained at several ages for the MP series between 0 05 MPa and 0 18 MPa and a higher variation was obtained 0 40 MPa at only one age for the SP series 5 3 2 3 Shear stress on saturated surfaces The cores with SP preparation were only tested at 38 days At this age the bond values observed for the SP series were lower
5. Barcelona d de de Acknowledgments i ACKNOWLEDGEMENTS AGRADECIMIENTOS Comienzo agradeciendo a las dos personas que hicieron posible esta tesis y en general esta experiencia De lo general a lo particular primero Antonio Aguado una persona con una visi n global excelente y segundo Alejando Josa detallista y meticuloso Juntos conforman un equipo magnifico que se complementa casi a la perfecci n Gracias por el tiempo paciencia y sabidur a que me han dedicado Agradezco la financiaci n brindada por el Ministerio de Educaci n y Ciencia a trav s del proyecto BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras Al Programa de FPU del Ministerio de Educaci n por la financiaci n para la realizaci n del doctorado AP2010 3789 A Aguado por la financiaci n al comienzo de la tesis a trav s del proyecto CTT 8062 Y al Instituto de Estructuras y Transporte de la Universidad de la Rep blica Uruguay por el constante apoyo recibido Al personal del Laboratorio de Tecnolog a de Estructuras en especial Jorge Cabrerizo por su ayuda durante la campa a experimental Mi recuerdo a Luis Agull que empapaba el ambiente con su alegr a A todos los compa eros de doctorado Andr Francisco Francesc Albert Nayara Ju Izelman Sandra Pau Ana Ahmed Amin Liao Nacho Catalina Luca Yohei Renan J lia Un agradecimiento especial a Sergio que estuvo siempre dispuesto y abierto
6. El hormig n de segunda fase fue proyectado por v a h meda completando el elemento estructural La dosificaci n utilizada para este tipo de hormig n se dise buscando que hormigones de ambas fases tuviesen caracter sticas mec nicas lo m s similares posibles Los testigos para estudiar la adherencia entre capas se extrajeron del muro un d a antes de la fecha prevista para su ensayo De esta forma stos tienen durante el mayor tiempo posible las mismas condiciones de curado que el resto del elemento Ya que se plane realizar ensayos a diferentes edades la extracci n Bi layer diaphragm walls Experimental and numerical analysis 116 Appendix 1 tambi n se realiz en diferentes edades Cuando el hormig n de segunda fase ten a 1 d a 5 dias y 34 d as de edad se realizaron extracciones de testigos en las zonas PF y PE Para cada edad se extrajeron 5 testigos de cada una de las zonas Se cuenta con series parcialmente incompletas para ambos tipos de ensayos La raz n principal es que algunos testigos tanto para el ensayo de corte como para el pull off y principalmente los extra dos en las edades mas tempranas se rompieron por el plano de uni n en el momento de realizar la extracci n o la perforaci n parcial Informaci n complementaria sobre el programa experimental se puede obtener en Segura Castillo amp Aguado de Cea 2012a alibraci n i PEU i i PF B i P
7. Fig 32 shows the bending moment envelopes for the cases with 1 2 and 4 spraying stages The negative local maximums and the moments developed in the intermediate stages where the second layer was sprayed Bi layer diaphragm walls Experimental and numerical analysis 66 Chapter 4 are also shown in the figure with black lines A circle indicates the depth of the sprayed concrete layer at the corresponding stage 0 0 0 1 1 1 3 2 2 3 E 34 4 4 44 3 5 1 243 6 5 1247 3 6 6 6 7 7 7 8 8 8 nn E 9 9 9 lt 10 10 10 300 2 298 6 307 6 a 11 11 1 0 a 12 12 12 13 NS 1S 13 NS 2S 13 NS 4S 14 Exc 12 50 14 Exe 7 75 144 Exc 3 75 as DT Supp Out 15 Exc 12 50 qe er Exe 7 75 157 ys 154 Supp Out 157 Exc 9 75 16 MP 16 MP 16 Exc 12 50 Depth of MP Supp Out Es z E z sprayed layer ad Depth of 17 Mi 18 i 18 sprayed layer 1B lt 5 a l 19 19 19 opine 20 a 20 b 204 o sprayed layer c 400 300 200 100 0 100 200 300 400 300 200 100 0 100 200 300 400 300 200 100 0 100 200 300 Bending Moments kN m Fig 32 Bending moment envelope for different spraying discretizations a 1 spraying stage b 2 spraying stages c 4 spraying stages It can be seen that in the case of a single spraying stage BL 60 10 BUs 4u 1S A see Fig 32a the area of bi layer work is small re
8. Roscoe K H amp Burland J B 1968 On the generalized stress strain behaviour of wet clay In Engineering plasticity pp 535 609 Cambridge Heyman and Leckie Sanhueza Plaza C X amp Oteo C 2007 Estudio Comparativo sobre Diferentes Modelos de C lculo Aplicados a la Construcci n de Muros Pantalla Revista de la Construcci n 6 1 13 27 Bi layer diaphragm walls Experimental and numerical analysis 108 References Saucier F Bastien J Pigeon M amp Fafard M 1991 A combined shear compression device to measure concrete to concrete bonding Experimental Techniques 15 5 50 55 Retrieved from http onlinelibrary wiley com doi 10 1111 1747 1567 1991 tb01214 x abstract Schanz T Vermeer P A amp Bonnier P G 1999 The hardening soil model formulation and verification In Beyond 2000 in computational geotechnics 10 years of PLAXIS pp 1 16 Rotterdam Balkema Schneebeli G 1981 Muros pantalla t cnicas de realizaci n m todos de c lculo Editores T cnicos Asociados Segura Castillo L amp Aguado A 2011 2011NT06 V2 An lisis final de la experiencia edificio Aprestadora in Spanish p 119 Barcelona Segura Castillo L Aguado A de la Fuente A amp Josa A 2013 Bi layer diaphragm walls Structural and sectional analysis Journal of Civil Engineering and Management Accepted for publication Segura Castillo L Aguado A amp Josa A 20
9. l J T f Fig 6 Experimental building a site plan b general cross section c detail of bi layer cross section The design of the experimental campaign was based on an uncoupled structure section analysis The structural analysis was performed using the Cypecad CYPE Ingenieros 2011 module for diaphragm walls Bi layer diaphragm walls Experimental and numerical analysis 24 Chapter 2 a FEM based program which considers soil structure interaction modelling the walls with FEM beam elements and the soil with a Winkler model The numerical simulation of the mechanical behavior of the composite sections of the Wall was performed with the model Analysis of Evolutionary Sections AES de la Fuente Aguado de Cea amp Molins 2008 de la Fuente Aguado de Cea Molins amp Armengou 2012 This model allows simulation of the non linear response of sections built with different materials concrete and steel and the structural contribution of the SFRC under tensile stress The Auxiliary Anchorage in Wall W35 was deliberately placed to cause flexural moments in the wall once the bi layer section had been constructed facilitating the analysis of the structural collaboration When the Auxiliary Anchorage was eliminated a bending increase in the wall occurred to redistribute the forces to the remaining anchorages and to the footing of the wall placing the bi layer cross sections under gre
10. 1934 1939 doi 10 1016 j buildenv 2005 06 023 Khoiri M amp Ou C Y 2013 Evaluation of deformation parameter for deep excavation in sand through case histories Computers and Geotechnics 47 57 67 doi 10 1016 j compgeo 2012 06 009 Kung G T C Juang C H Hsiao E C L amp Hashash Y M A 2007 Simplified Model for Wall Deflection and Ground Surface Settlement Caused by Braced Excavation in Clays Journal of Geotechnical and Geoenvironmental Engineering 133 6 731 747 doi 10 1061 ASCE 1090 024 1 2007 133 6 731 Kung G T C 2009 Comparison of excavation induced wall deflection using top down and bottom up construction methods in Taipei silty clay Computers and Geotechnics 36 3 373 385 doi 10 1016 j compgeo 2008 07 001 Kurk F amp Eagan P 2008 The value of adding design for the environment to pollution prevention assistance options Journal of Cleaner Production 16 6 722 726 doi 10 1016 jclepro 2007 02 022 Li K Ju Y Han J amp Zhou C 2008 Early age stress analysis of a concrete diaphragm wall through tensile creep modeling Materials and Structures 42 7 923 935 doi 10 1617 s11527 008 9432 4 Bi layer diaphragm walls Experimental and numerical analysis 106 References Lim A Ou C Y amp Hsieh P G 2010 Evaluation of clay constitutive models for analysis of deep excavation under undrained conditions Journal of GeoEngineering 5 1 9 20 Retrieve
11. Carrubba amp Colonna 2000 Costa Borges amp Fernandes 2007 Ou amp Lai 1994 have evaluated the forces and stresses produced on the walls It should be mentioned that use of the elastic perfectly plastic Mohr Coulomb model means that the soil has to be discretized into several horizontal layers and its elastic properties have to be changed as the depth increases before the model responds to any increase in the soil modulus of elasticity due to increased Luis Segura Castillo Structural and sectional analysis 43 vertical pressure Khoiri amp Ou 2013 The hardening soil model HS Schanz et al 1999 was therefore chosen as it models the entire ground in the study with only one set of parameters The numerical simulation of the mechanical behaviour of the composite sections of the wall was performed with the model Analysis of Evolutionary Sections AES de la Fuente Aguado de Cea et al 2012 This model simulates the non linear response of sections built with different materials concrete and steel and the structural contribution of the SFRC when subjected to tension In the AES model the concrete sections are discretized in layers of constant thickness see Fig 20a whereas steel rebars are simulated as concentrated area elements In this study the procedure to design the reinforcement of the concrete wall followed the basic design principles for traditional reinforced concrete presented in EN 20
12. una pantalla continua previamente fresada y la b squeda de un ensayo factible de ser utilizado para el control rutinario de dichas pantallas Las siguientes conclusiones pueden ser extra das La resistencia al corte con la superficie tratada con fresado aumenta con la edad del hormig n de segunda fase ajust ndose a las f rmulas de maduraci n Para el mismo ensayo la preparaci n de la superficie con Epoxi en obra presenta gran dispersi n de resultados con un CV de hasta 57 en el peor de los casos Debido a las caracter sticas del elemento propuesto principalmente el espesor de la segunda capa y las variaciones del plano de interfase y del paramento interior con el plano medio de la pantalla el ensayo pull off registra una gran dispersi n de valores No se puede por lo tanto en las condiciones en las que se realiz esta experiencia recomendar la utilizaci n de este ensayo para el control rutinario de la adherencia A la vista de los resultados obtenidos para la utilizaci n en muros bi capa o grandes superficies en general salvo que se realicen controles rigurosos en los tiempos de disposici n y ejecuci n del puente de adherencia y del proyectado del hormig n no es aconsejable la utilizaci n de productos epoxi Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 125 Debido a la gran dispersi n obtenida en las distintas series para el ensayo pull off consid
13. Bi layer diaphragm walls Experimental and numerical analysis 56 Luis Segura Castillo Chapter 3 Seven Deadly Sins Wealth without work Pleasure without conscience Science without humanity Knowledge without character Politics without principle Commerce without morality Worship without sacrifice Mahatma Gandhi CHAPTER 4 Parametric study of construction processes ABSTRACT The bi layer diaphragm wall is a new type of slurry wall designed to improve watertightness and to counter leakage problems These walls consist of two bonded concrete layers the first a conventional Reinforced Concrete RC diaphragm wall and the second a sprayed Steel Fibre Reinforced Concrete SFRC layer with a waterproof additive Here we analyse and quantify the influence of different construction process parameters on the effectiveness of the bi layer diaphragm wall technique Thirty numeric simulations were conducted with an uncoupled structure section analysis placing special emphasis on the SFRC layer contribution The results show that in all cases the main flexural strength is provided by the RC layer with a secondary flexural contribution between 8 and 15 by the sprayed SFRC layer Using satisfactory spraying sequences detailed herein a reduction in the steel reinforcement of the RC layer can be obtained in every structural configuration and construction sequence reaching a maximum percentage reduction of 7 0 of the total bendi
14. En la Fig 69 se incluye l nea de mejor ajuste para la totalidad de los valores es decir sin diferenciar por edad de ensayo A pesar de la dispersi n mencionada en el apartado anterior se observa una leve correlaci n para los ensayos realizados en la superficie con epoxi Que ambos ensayos brinden resultados correlacionados induce a pensar que la dispersi n obtenida es producida por las dispersiones en el valor de Bi layer diaphragm walls Experimental and numerical analysis 124 Appendix 1 la adherencia y no a que son producto del ensayo utilizado confirmando lo planteado en el apartado A1 4 1 2 referido a la variabilidad de la respuesta del epoxi cuando es utilizado en esta aplicaci n 0 8 07 y 0 3815x 0 0136 O gt R 0 73 0 6 O 0 5 a I 0 4 r a 0 3 AY 3 O O 3 0 2 A 301 30 OPE 73 OPE_T2 i J 0 0 5 1 1 5 2 Resistencia a corte t MPa Fig 69 Resistencia a tracci n vs Resistencia a corte para preparaci n con adhesivo Epoxi Para las parejas relacionando las resistencias obtenidas para la superficie fresada la correlaci n entre ambos ensayos se pierde pudi ndose observar simplemente una zona de concentraci n de valores A1 5 CONCLUSIONES El presente estudio experimental analiza los muros pantalla bi capa particularmente la adherencia alcanzada entre el hormig n de la segunda capa colocado mediante proyectado y el de la primera capa perteneciente a
15. Fig 46 Failure mechanisms in the test Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 87 This type of failure took place in some cases for instance in the example shown in Fig 47a In other cases failure occurred simultaneously on both planes Fig 47b As this is an abrupt type of failure the surface on which it first took place could not be determined The chip was measured shown as distance s in Fig 47a from its lower edge perpendicular to the direction of the load to the uppermost point of the core When failure took place largely outside the bond plane of the two concretes s gt 3 cm the test was declared null and the result was dismissed Fig 47 Atypical failure a stone chip protruding on one side b on two planes In order to reduce the influence of these secondary mechanisms failure could be induced on the bond surface producing notches on the edges of the bond surface which leads to failure only along this plane a strategy that has been validated in previous studies Mirsayah amp Banthia 2002 5 3 2 2 Shear stress depending on age Fig 48 shows a graph of the average value and represents the standard deviation for each age of the MP and SP series 2 00 180 _____ y 0 2127In x 0 8549 1 60 TIR OS TM 1 40 4 1 20 A LL 1 00 LL LL 0 80 d e 0 60 0 40 H SP OMP 020 los omel
16. M Su The main flexural strength is provided in all cases by the strength of the first layer with conventional bars reinforcement while the second layer with SFRC provides a secondary flexural contribution Secondly it can be seen from Table 17 that the spraying sequence is a relevant parameter in the design of bi layer walls Walls with different spraying sequences but with the same final structural configuration and construction sequence lead to widely different steel variations For example only a significant steel reduction 13 5 kg could be obtained for the case with SD B in the first group This behaviour is studied in further detail in the following sections It is worth mentioning that for all the final structural configurations and construction sequences under study there is at least one bi layer alternative where a steel reinforcement reduction greater than 10 kg is obtained The largest reduction 21 7 kg is reached in the case of a TD construction sequence and 60 cm first layer thickness BL 60 10 TD 2u 2S B It represents a percentage reduction of 7 0 of all the longitudinal steel reinforcement and of 45 1 in terms of the extra steel reinforcement as defined at the end of section 4 2 1 Finally it can be seen that the maximum displacements are closely related to the construction sequence that is followed with only minor reductions smaller than 0 6 mm if the bi layer type is used At the same time it may be
17. MP case ceeccccesceceeeceeseeceesaeceeneeceeeeeeeeeaeceeaeecenees 89 51 Milling marks On tested Cores t0 2 5 yesveasoersespestun eae cite debe ccliavl jeden cascadas ceca ets Sevres 90 52 Failure plane elena daria dies 91 53 Influence of failure angle on shear strength ooconnccnnnnnnccnonoconnconnnannconnnono cana con nono nonnocnnncnnncnnnnnnacnn noo 91 54 Influence of the strength of base concrete on shear strength MP case oooconocccocccocnnocnnacnnonncanaconnno 92 55 Defectos usuales en pantallas continuas a Diferencias de posici n entre bataches adyacentes b P rdidas entre juntas durante construcci n c P rdidas que aparecen ya en servicio o 112 56 Esquema de la soluci n por muro Bi Capa Vista general y vista de una Secci nN e e oeee 113 57 Detalle de las pantallas a Plano del sitio de obra b corte lateral oonooonncccninanincccnnocaconacanancnonnss 115 58 Esquema de posiciones de extracci n de testigos para el ensayo de corte y de realizaci n de ensayos pull off in situ para las distintas preparaciones e imagen de la pantalla luego de realizadas las EXACCION Si A A A bi 116 59 Ensayo de Corte LCB a Esquema del dispositivo b Configuraci n del ensayo 117 60 Ensayo Pull off a Esquema del ensayo b Configuraci n del ensayo 118 61 Resistencia a corte vs edad del hormig n de 2 fase ooooconoccconccoccconn
18. Near Surface Mounted among others Bonaldo Barros amp Lourenco 2005 Bond strength depends mainly on interface adhesion friction aggregate interlock and time dependent factors Momayez Ehsani Ramezanianpour amp Rajaie 2005 An essential requirement relates to the development of full bond strength over time between the reinforcement and the base layer Talbot Talbot Pigeon Beaupr amp Morgan 1994 in reinforcement with sprayed concrete and Delatte Delatte Williamson amp Fowler 2000 in reinforcement for bridges both studied bond durability and maturity at increasing ages The values obtained for the strength of the bond depend strongly on the chosen test method Momayez et al 2005 Several authors have performed different studies which on the one hand describe and classify the methods and on the other hand compare the results they obtain Abu Tair Rigden amp Burley 1996 Simon Austin Robins amp Pan 1999 Julio Branco amp Silva 2004 Momayez et al 2005 The slant shear test Wall amp Shrive 1988 has become the most widely accepted test and has been adopted by several international regulations as the test for assessing the bond between resinous repair materials and the base concrete Abu Tair et al 1996 However there is no agreement among researchers with regard to the suitability of non resinous materials Momayez et al 2005 The lack of consensus over any one test or another may be
19. in the cases shown in chapter 2 10 m high walls on heterogeneous soil with RC layer thickness of between 35 cm and 45 cm and more than 20 in chapter 3 20 m high walls on sandy soil with RC layer thickness of between 55 cm and 60 cm Overall flexural design model structural and sectional level A complete flexural design method was presented The method is based on a FEM structural model developed with PLAXIS a commercial geotechnical oriented FEM software and the AES sectional model a numerical model for the analysis design and checking of composite sections developed in UPC The complete reinforcement bars of the first layer and fibres in the SFRC layer can be designed and checked using this method The smallest moments in the simple section and the collaboration of the SFRC layer afford a reinforcement reduction 2 1 and 2 5 in the chapter 3 cases and up to 7 if an adequate spraying sequence is used in the chapter 4 cases In both chapters 20 m high walls on sandy soil with RC layer thickness of between 55 cm and 60 cm were used 6 3 5 Influence of the different constructions processes related to this type of walls It is feasible for all final structural configurations and construction sequences to achieve a reinforcement steel reduction of the first layer by taking full advantage of the strength added by the SFRC layer Luis Segura Castillo Conclusions and future perspectives 99 6 3 6 6 3 7 6 4 U
20. n si se supone que la temperatura media del hormig n para el transcurso de tiempo estudiado es constante Esta es una hip tesis razonable si se considera que el elemento es de un espesor peque o y est en contacto con el terreno 2 0 y 0 2127In x 0 8549 18 R 0 9754 a 16 A a A 514 o A ys GJ A 812 0 e gt 1 0 Oo A o E 0 8 PF OPE 0 6 1 10 100 Edad del hormig n de refuerzo d as Fig 61 Resistencia a corte vs edad del hormig n de 2 fase 0 8 0 7 0 6 0 5 0 4 0 3 0 2 0 1 I t1 2 3 t2 6 t3 35 Edad del hormig n de refuerzo d as Desviaci n estandar MPa Fig 62 Desviaci n est ndar obtenida para las distintas edades Bi layer diaphragm walls Experimental and numerical analysis 120 Appendix 1 A1 4 1 2 Preparaci n con puente de adherencia epoxi En la primer edad t 3 d as solo se pudo ensayar un testigo por lo que no se puede evaluar la dispersi n para este tiempo Para las otras dos edades consideradas se obtuvieron dispersiones muy diferentes En la segunda edad t 6 d as se registr la mayor dispersi n 0 68 MPa y en la tercer edad una dispersi n mucho menor 0 11 MPa Se adjudica la mayor variabilidad de los resultados del caso PE a las dificultades de aplicaci n del v nculo de adherencia inherentes de la obra Luego de aplicar el producto se debe esperar una hora antes de poder colocar el hormig n de segunda etapa A su v
21. required for the construction of the complete systems the maximum displacement registered the final thickness of the system and its waterproofing if any Luis Segura Castillo Structural and sectional analysis Table 15 Comparison of different waterproofing systems 53 Mono layer Bi layer ML 60 ML 50 ML 50 BLoo 10 BLss io BL oo wio BLssw10 DC WML Unif Unif Opt Opt Cast layer m m 12 00 12 00 12 00 12 00 11 00 12 00 11 00 Concrete 3 as Sprayed layer m m 0 63 1 25 1 25 1 25 1 25 Total m m 12 00 12 00 12 63 13 25 12 25 13 25 12 25 Steel Cast layer kg m 391 391 391 382 381 382 381 ee Sprayed layer kg m 31 31 18 18 weight Total kg m 391 391 391 414 412 400 399 Maximum displacement mm 17 3 17 3 17 3 17 6 18 1 17 6 18 1 Final thickness cm 60 85 65 70 65 70 65 Waterproof system NO YES YES YES YES YES YES corresponds to the volume of the mortar layer in the MLso WML system the theoretical thickness is considered without taking accidental deviations into account the consumption of extra materials of a different class is not considered in this system The different materials are grouped below in accordance with their class Thus the volume of the two types of concrete RC and SFRC and the mortar used in the WML system are grouped under the heading concrete and the conventional steel bars used in the RC layer and the
22. se observ un comportamiento monol tico a nivel elemento para las pantallas BL experimentales Se obtuvo un consumo final de materiales similar entre pantallas BL y la consideraci n conjunta de una pantalla ML m s un sistema impermeabilizante externo Considerar los costos tecnol gicos conlleva un costo constructivo mayor para las pantallas BL Sin embargo sta es a n una opci n interesante bajo consideraciones particulares como limitaciones del espacio subterr neo interior o si se desean evitar costos continuos de mantenimiento Bi layer diaphragm walls Experimental and numerical analysis vi Summary En t rminos generales se puede decir que la investigaci n aqu presentada sienta las bases para el desarrollo de la t cnica de muros pantalla bi capa la cual es una soluci n prometedora para el problema de las filtraciones en pantallas No obstante son necesarios m s estudios para poder usar plenamente este tipo de pantallas de forma habitual e g estudios de sostenibilidad detallados evaluaci n del riesgo de desprendimiento de la segunda capa capacidad impermeable y sobre todo m s ensayos experimentales a escala real Luis Segura Castillo Table of contents vii TABLE OF CONTENTS ACKNOWLEDGEMENTS AGRADECIMIENTOG sssscccccccssssssssssceeecccccssssssseceececcccsssscesseccecccccsssssceescescascssssscessoeoess I SUMMARY A O Il A A O V TABLE OF CONTENT Ssiesecicescceseccisccdiscceceascsssesicescctescdseseds
23. structural elements like base slabs to avoid leakages in the connections Luis Segura Castillo References 101 REFERENCES Abu Tair A Rigden S amp Burley E 1996 Testing the bond between repair materials and concrete substrate ACI Materials Journal 93 6 553 558 Retrieved from http www concrete org PUBS JOURNALS OLJDetails asp Home MJ amp amp ID 9861 ACI Committee 116 2000 ACI 116R 00 Cement and Concrete Terminology 1 73 ACI Committee 212 2010 ACI 212 3R 10 Report on Chemical Admixtures for Concrete Vol 1 ACI Committee 544 2002 ACI 544 1R 96 State of the Art Report on Fiber Reinforced Concrete Vol 96 p 66 American Concrete Institute AFTES W G N 6 1996 AFTES recommendations on fibre reinforced sprayed concrete technology and practice Tunnelling and Underground Space Technology 11 2 205 214 doi 10 1016 S0886 7798 96 90088 7 Aguado A Blanco A de la FUENTE A amp Pujadas P 2012 Manual Sobre el Hormig n con Fibras In Spanish p 224 CEMEX UPC ASTM 2009 D4541 Standard Test Method for Pull Off Strength of Coatings Using Portable Adhesion Testers Austin S 2002 Sprayed concrete technology p 300 Chapman amp Hall Austin Simon amp Robins P J 1995 Sprayed Concrete Properties Design and Application Whittles Publishing Services Austin Simon Robins P amp Pan Y 1995 Tensile bond testing of concrete repairs Materials a
24. the walls to the axis of symmetry of the model was used for the upper slabs A stiffness of EA 1 6510 kN m equivalent to a 50 cm thick massive slab also with an equivalent support length of 10 m was used for the bottom slab Geogrid structural elements and node to node anchor elements were used to model the body and the free length respectively of the ground anchors The following properties were used Initial tensile load 50 RN m for S A 1 and 2 and 100 kN m for LA 1 and 2 Horizontal distance 5 0 m for S A 1 and 2 and 2 5 m for LA 1 and 2 The rest of the properties are equal for both types of anchorages Total length 20 m Bulb length 14 m Cross section area 450 mm Elastic modulus 200 KN mm and Angle 30 A fibre content of 25 kg m was used in the SFRC The post cracking behaviour of the SFRC may be defined by the expressions given in de la Fuente Escariz de Figueiredo Molins amp Aguado de Cea 2012 The nominal cover used for the RC bars was 70 mm 3 4 STRUCTURAL RESULTS The displacement plot of the three wall types corresponding to the final stage Anc Ourt are shown in Fig 23 The upper part of the plot is enlarged for clarity In general terms the displacement of each wall type is similar with differences in the maximum displacement value of less than 0 8 mm 4 6 and within the order of magnitude of displacements of the reference case Carrubba amp Colonna 2000
25. 2 anchorages in the Bottom Up cases and 2 slabs in the Top Down case In the latter case after the base slab is in place no further bending moments are evident 0 a 1 a 2 2 ps 3 3 4 4 lo s 5 i y 6 3 2 7 pi 1 a 4 s E 3 nd o F 107 NM a a 11 pa a 12 eri B7 cs Bua a 44 14 Exc 3 75 Ll T 15 Exc 6 75 5 thy E y a 17 Exc 12 50 a 18 R e El as 1a 194 ys Bj 600 400 200 0 200 400 600 400 200 0 200 400 600 600 400 200 0 200 400 Bending Moments kN m Fig 37 Bending moment envelope for different construction sequences a Bottom Up with struts b Bottom Up with anchorages c Top Down It can be seen that the BUa case see Fig 37b shows a similar shape in the bending moment diagrams as the BUs case However the values are significantly higher because the upper part of the wall has larger displacements as the supports are more flexible the anchorages have lower stiffness than the struts The embedded section of the wall is therefore under greater strain increasing the bending moments along the wall Behaviour quite unlike the previous two cases resulting in a different diagram shape can be seen in the case with the TD construction sequence see Fig 37c As the slabs have a significantly higher stiffness than the temporal supports and are built during the excavation process they apply greater reactive forces than the
26. 200 400 600 400 200 0 200 400 Bending Moments RN m Fig 39 Design and ultimate bending moments envelope examples a efficient design b inefficient design c large increase in bending moments after spraying d Top Down case Fig 39a shows a case BL 60 10 BUs 4u 28 B where the maximum design moments are relatively similar ion Gie ME MG MS MS to the ultimate moments for each type of cross section i e u xta and Hu a The moments that are recorded up until the second layer was sprayed are resisted by the M gt and the subsequent moment increase is resisted by the My It can be considered an efficient design in the sense that in the ULS the entire strength of the materials is needed to resist the design moments both in the simple cross section and in the compound cross section Ss G In a case where the maximum M4 is equal to the maximum Mj see case BL 60 10 BUs 4u 2S C Fig 39b the M v resisted both design moments In contrast to the previous one this design may be considered inefficient because even though the second layer increases the strength of the cross section this increase is not needed to cover the design moments Therefore it can be said that spraying the SFRC layer is a necessary but not a sufficient condition the spraying sequence must also be taken into account to allow a reduction in the steel reinforcement Fig 39c shows a case BL 60 10 BUa 2u 2S B where there is a l
27. A A Balkema Publishers Lisse The Netherlands Brown A J amp Bruggemann D A 2002 Arminou Dam Cyprus and construction joints in diaphragm cut off walls G otechnique 52 1 3 13 doi 10 1680 geot 2002 52 1 3 Bryson L S amp Zapata Medina D G 2012 Method for Estimating System Stiffness for Excavation Support Walls Journal of Geotechnical and Geoenvironmental Engineering 138 9 1104 1115 doi 10 1061 ASCE GT 1943 5606 0000683 BS 8102 1990 Code of practice for protection of below ground structures against water from the ground p 40 British Standard Institution BS 8102 2009 Code of practice for protection of below ground structures against water from the ground British Standard Institution Calvello M amp Finno R J 2004 Selecting parameters to optimize in model calibration by inverse analysis Computers and Geotechnics 31 5 410 424 doi 10 1016 j compgeo 2004 03 004 Carrubba P amp Colonna P 2000 A comparison of numerical methods for multi tied walls Computers and Geotechnics 27 117 140 Retrieved from http www sciencedirect com science article p11 S0266352X00000070 Celestino T B amp Ishida A 2009 ACTIVITIES OF THE ITA WORKING GROUP ON SPRAYED CONCRETE USE In ECI Symposium Series Ed Shotcrete for Underground Support XI Retrieved from http dc engconfintl org shotcrete Chan R VV M Ho P N L amp Chan E P VV 1999 Report on Concrete Adm
28. ARAS cava AAA GAS Gatita 112 A1 1 3 Adherencia entre NOLMiGOne c ccccccccceesessscesecsnsssssssesecsssesaascesecesesssassessesseessasssesscsseassassesseneeaaes 113 OBJETIVOS EE E AA seers eadeleeevicbedeavsideceatteeiunleageaverdeceass 114 METODOLOG A raras teta tn asain a ou tesanouvaaseonesta sentzatesauoncmonasta Rao A Da Ne CORTES 114 A1 3 1 PrOJFrAMA CXPEIMEN Alina alinea dani 114 A1 3 2 Zonas de extracci n de testigos cirereta iadenin ia i aia aderai iaaa ia iin 116 A1 3 3 ENSOVONOCO eri ie iiaia mm AAEE EA TETO E O inte 117 A1 3 4 ENSOVO PU Ou tai 117 RESULTADOS ANAL Sii sis 118 A1 4 1 Resultados principales del ensayo de corte u cessessssscecccessessscsecsceeseseascecsceesesseesecsseeseaasseseceeeene 118 A1 4 1 1 Preparaci n por fr sad iii iii picado ia ile 118 A1 4 1 2 Preparaci n con puente de adherencia epoxi cccsccssccsstecsseeeeseesseesseeceseceseeseaeessnessasesseeseaeesseeseatenee 120 A1 4 2 Resultados pulFO Fic a aaa 120 A1 4 3 Relacion Corte pull Off ic em A A sine dc RE 123 A A ON 124 AGRADECIMENTO S e a EES a EETA oia 125 ANEXO RESULTADOS EXPERIMENTALES DEL ENSAYO PULL OFF scccceceeeeeesnececeeceeneaaececececesseaaeeeceeeeeaaeceeeeeseeneaaees 126 Bi layer diaphragm walls Experimental and numerical analysis Table of contents Luis Segura Castillo List of figures xi LIST OF FIGURES Fig 1 Typical defects of diaphragm walls in joints between panels cece ceec
29. Agull Fit Aguado de Cea amp Rodr guez Barboza 2001 and the experience of the concrete manufacturer MEYCO MS 685 OPTIMA 209 nano silica was used as a superplasticizer The characteristics of the metallic fibres that reinforce the structural composition of the composite element were as follows Length 35 mm diameter 0 55 mm with hooked ends brand name DRAMIX RC 65 35 BN The required thickness of the second layer was 10 cm but due to the intrinsic irregularity of the spraying system layer thicknesses ranging from 9 cm to 17 cm were detected in the subsequent extraction of cores An Aliva 503 robot was used to for concrete spraying once the base concrete was 84 MP and 86 EP and SP days old After spraying the surface was kept wet for a whole day During the spraying of the second Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 83 layer two casts were filled with the same concrete and the procedure outlined in UNE EN 14488 1 UNE EN 14488 1 2006 was followed cores were extracted from the casts and were used to determine their compressive strength UNE EN 12390 3 UNE EN 12390 3 2003 Table 20 Types of surface preparation Name of preparation Milled Surface MP Surface with epoxy EP Saturated Surface SP Mechanical treatment Surface milling Surface milling Surface milling Surf ist Saturated and left to dry one day Saturated and left to dry one day Saturated during second s
30. Bonaldo et al 2005 Este comportamiento es necesario en el campo de la reparaci n y refuerzo de estructuras de hormig n donde la pr ctica com n consiste en en primer lugar incrementar la rugosidad de la capa base en algunos casos aplicar un puente de adherencia y o conectores de acero y posteriormente colocar la capa de refuerzo J lio et al 2006 Ejemplos habituales de esta aplicaci n incluyen puentes construidos en varias etapas o t cnicas de refuerzo de pavimentos Delatte Jr et al 2000 o t cnicas m s complejas como la NSM Near Surface Mounted Bonaldo et al 2005 Entre otros autores Talbot et al 1994 en refuerzo con hormig n proyectado y Delatte et al 2000 en refuerzo para puentes estudiaron la durabilidad y maduraci n del v nculo con la edad Bi layer diaphragm walls Experimental and numerical analysis 114 Appendix 1 Los valores obtenidos de la resistencia del v nculo dependen fuertemente del m todo de ensayo elegido Momayez et al 2005 Varios autores han realizado diferentes estudios que por un lado describen y clasifican los m todos y por otro relacionan los resultados por ellos obtenidos Abu Tair et al 1996 Simon Austin et al 1999 Julio et al 2004 Momayez et al 2005 El slant shear test Wall amp Shrive 1988 se ha convertido en el ensayo mas ampliamente aceptado y ha sido adoptado por varias normas internacionales como ensayo para evaluar la adherencia de materiales
31. Geoenvironmental Engineering 136 7 985 994 doi 10 1061 ASCE GT 1943 5606 0000299 Whittle A J 1987 A constitutive model for overconsolidated clays with application to the cyclic loading of friction piles Massachusetts Institute of Technology Wong I 1997 Experience with waterproofness of basements constructed of concrete diaphragm walls in Singapore Tunnelling and Underground Space Technology 12 4 491 495 Retrieved from http linkinghub elsevier com retrieve pii S088677989800008X Xanthakos P P 1979 Slurry walls p 622 New York McGraw Hill Zollo R F 1997 Fiber reinforced concrete an overview after 30 years of development Cement and Concrete Composites 19 2 107 122 Retrieved from http linkinghub elsevier com retrieve p1i S0958946596000467 Bi layer diaphragm walls Experimental and numerical analysis 110 References Luis Segura Castillo Se abre el tel n Un cubo de hormig n C mo se llama la obra Atracci n fatal APPENDIX 1 Early ages concrete to concrete bond strength assessed through shear and pull off tests ABSTRACT Leakage represents a widespread problem in diaphragm walls built under certain conditions such as enclosures in water bearing ground An innovative structural element typology is proposed referred to as a bi layer diaphragm wall Its two layers are poured and sprayed respectively in two phases the first layer is a standard reinforced concrete diaphragm wall wh
32. Nook Qu Run o 19 18 17 16 15 b o uo un sb wKH KO E 10 E Q 11 5 O 12 13 14 15 16 1 Me 184 BL 60 10 19 BL 55 10 20 18 16 14 12 10 8 6 4 2 0 Horizontal Displacements mm Fig 23 Wall displacements Anc Out stage for the three walls Displacements at depths of between 5m to 20m of the walls with a RC width of 60 cm MLeo and BL o 10 are practically identical The reduced influence of the second layer is on the one hand due to the Bi layer diaphragm walls Experimental and numerical analysis 46 Chapter 3 stiffness of the two types of walls which are the same at depths of between 12 5m and 20 0m On the other hand increased stiffness at depths of between 6 0m and 12 5m following spraying of the SFRC layer is noted during the final excavation stages after most of the soil pressure had been already mobilized The displacements of the BL55 10 wall type are slightly higher than the previous ones This behaviour which coincides with data reported in Segura Castillo Aguado amp Josa 2013 is due to the fact that the stiffness of the RC layer determines the overall displacement behaviour Larger displacements than in ML o wall at depths of between 0 0 m and 5 0 m can be seen in the BL o 10 wall Although it might appear contradictory this is reasonable because the flexural stiffness of the bi layer wall increases after spraying of
33. Once this wall attains the necessary strength subsoil in contact with the wall within the perimeter is excavated and removed and the second layer this time of sprayed Steel Fibre Reinforced Concrete SFRC and a waterproof additive is applied both layers form the compound cross section see Fig 27b Compound Section que ai a o pe Sa oa am Sa bm a oy Bond plane Area with sprayed concrete Area as yet H unsprayed 2d layer sprayed SFRC Ground Simple Section _ I 1st layer conventional RC os Fig 27 Bi layer diaphragm walls a general scheme b compound cross section c simple cross section and d spraying of an experimental wall The main objective of this paper is to analyze and quantify the influence of different construction process parameters in the efficiency of the bi layer diaphragm wall technique measured in terms of reduction in the reinforcement and in displacement This paper is part of an experimental and theoretical study of bi layer diaphragm walls structured into four main areas a Structural level analysis Segura Castillo Aguado de la Fuente amp Josa 2013 Segura Castillo Aguado amp Josa 2013 b Sectional level analysis Segura Castillo Aguado de la Fuente et al 2013 c Bonding between layers Segura Castillo amp Aguado de Cea 2012a and d General design and optimization This paper sets out the basis for the fourth of these aforementioned areas
34. S 0 80 HH HHH 7 0 60 H HH o 5 0 40 ane HH AHM OOM XVM 0 20 A 0 00 1 10 100 Age of 274 phase concrete days Fig 50 Direction of stress with regard to milling MP case According to these results there appears to be no connection between the milling direction and the direction in which the stress is applied in terms of the strength of the specimen Bi layer diaphragm walls Experimental and numerical analysis 90 Chapter 5 Examining the test specimens it can be observed that the failure plane to a great extent follows the bond plane and the original milling marks are still visible in many of the specimens Fig 51 Fig 51 Milling marks on tested cores Among the factors affecting bonding the milling direction can mainly be associated with aggregate interlock It is believed that in the same way as the friction mechanism this mechanism only contributes relevant strength where normal compression on the bond plane is able to mobilize it 5 3 2 6 Shear angle The first phase of concrete is cast on the ground and the second phase is sprayed It is therefore extremely difficult if not impossible for the surfaces of both concrete phases to be plain and parallel to each other This makes it impossible to extract a core with an axis that is perpendicular to the bond plane between the two concretes which is in all cases the plane of failure It was investigated whether the failure angle has any influenc
35. Se analiza adem s la influencia de las condiciones de contacto para lo que se ha analizado tambi n la alternativa de realizar el v nculo mediante una capa de imprimaci n con resina epoxi Todo ello es estudiado experimentalmente en una actuaci n real en un edificio A1 3 METODOLOG A A1 3 1 Programa experimental La tipolog a de pantallas descrita fue utilizada en un edificio ubicado en Barcelona Para el prop sito de este art culo se extrajeron testigos de las pantallas experimentales para la realizaci n de ensayos de corte y a su vez se realizaron ensayos pull off in situ como se describe a continuaci n En la Fig 57a se presenta el plano de la planta de la obra Las pantallas corresponden a todo el per metro de la obra habi ndose construidos por los m todos usuales Por otro lado en dicha figura se presentan las distintas zonas utilizadas para analizar las condiciones adherentes con la siguiente nomenclatura PF Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 115 Superficie Fresada PE Superficie con epoxi En la Fig 57b se muestra un esquema de la vista lateral de los mismos La primera fase de las pantallas Bi Capa est constituida por una pantalla convencional de hormig n armado con una resistencia a compresi n a 28 d as de f 30 MPa Al finalizar la excavaci n se realiz el fresado de la pared expuesta con el objeto de por un lado regula
36. The parameters under study are grouped into two categories a specific bi layer diaphragm walls characteristics i e number of spraying stages depth of sprayed concrete layer and b general diaphragm walls and construction characteristics i e wall thickness construction sequence final structural geometry Luis Segura Castillo Parametric study of construction processes 59 Many studies have reported on the parametric analysis of deep excavations studying the parameters of the second of the aforementioned categories The studies mainly involve two approaches analysis of a comprehensive case history database e g Clough amp O Rourke 1990 Long 2001 Moormann 2004 Wang Xu amp Wang 2010 and numerical analysis based on models calibrated against well documented case studies e g Bose amp Som 1998 Bryson amp Zapata Medina 2012 G T C Kung et al 2007 G T C Kung 2009 The main focus of these studies is on wall and ground movements due to their importance in the prediction of damage to adjacent buildings Thirty numeric simulations of diaphragm walls with varied parameters were run with an uncoupled structure section analysis to fulfil the objective Besides the displacements the structural response was also analysed focusing on the bending moments with special emphasis on the SFRC layer contribution 4 2 METHODOLOGY 4 2 1 Model description A 2D Finite Element Model FEM developed in PLAXIS was
37. a 16 0 cm que es en algunos casos un aumento considerable si se compara con los usualmente utilizados en los trabajos de referencia Simon Austin et al 1995 Bonaldo et al 2005 Delatte Jr et al 2000 J lio et al 2005 Momayez et al 2005 Talbot et al 1994 que van hasta 10 cm como m ximo Por otro lado por el sistema de construcci n de los muros pantalla bi capa es dif cil controlar los ngulos que forman el paramento la uni n entre capas el eje de perforaci n del testigo y el eje de esfuerzo del dispositivo de carga En la Fig 65 se observa un esquema de una secci n del muro La primera capa de hormig n es formada por el sistema tradicional de construcci n de los muros continuos Por lo tanto la forma de la superficie exterior queda determinada por la forma en la que se ha escavado el terreno La homogeneidad del plano exterior depende en gran medida de la maquinaria con la que se ha realizado la excavaci n y del tipo de suelo en el que esta se inserta Dependiendo de estas variables es com n en mayor o menor medida el desprendimiento de parte del suelo de los paramentos de la excavaci n dando lugar luego del hormigonado a la formaci n de barrigas En la cara interior del muro una vez expuesta luego de realizada la excavaci n interior es com n la realizaci n de un fresado de homogenizaci n en donde se eliminan estas barrigas De todos modos a n pueden permanecer variaciones m s suaves a lo lar
38. acceptable tests test age surface preparation mean bond stress calculated by means of formula 7 standard deviation and finally the direction with regard to milling used in the tests These results are examined in the following sections Table 22 Shear test results Series Number of Cores extracted debonded on Age Surface Tm Standard Milling direction extraction rejected tests acceptable tests days preparation MPa deviation MPa FV FH FO MP 2 5 2 0 3 2 MP 1 04 0 160 1 1 1 EP 3 5 4 0 1 3 EP 1 43 1 0 0 MP 6 5 1 0 4 6 MP 1 18 0 181 2 2 0 EP 6 5 2 0 3 6 EP 1 18 0 681 1 1 1 MP 35 5 1 2 2 35 MP 1 63 0 046 1 1 0 EP 35 5 0 0 5 35 EP 1 04 0 109 2 2 1 SP 38 4 0 1 3 38 SP 1 01 0 401 5 3 2 1 Types of failure As shown in Fig 45 brittle failure was detected in all cases Once the maximum load P was reached there was an immediate drop in shear strength Load P was used in formula 7 to assess the shear stress on the bond plane The recorded displacement corresponds to the displacement of the piston by the loading press which is therefore also affected by deformation or movements of the whole device in addition to any deformation of the concrete under shear stress The graph shows a first non linear stretch which becomes progressively stiffer as deformation increases So that the specimen moves in unison with the clamp tension should be transferred between them dispersed Bi layer diaphragm walls Experimental and nume
39. and numerical analysis viii Table of contents 26 2 3 OPtlMUM Mr ii AAA immi caga 37 2 6 3 GOMBOTISON vaste aaa teens caesssices ANN OT 38 2 7 CONCLUSIONS a escocesa NO 39 28 sACKNOWLEDGEMENTS wsacccscic cit et SS 40 CHAPTER 3 STRUCTURAL AND SECTIONAL ANALYSIS ssscccccccccssssssssceescccccsssssssceeeeccecessssseseceesccecesssseees 41 3 1 ANTRODUCTION si em e a a a a aea aaa a raaa a eaaa aa aa naaa aeaaea a Aars 42 3 2 lt METHODOLOGY AA A 42 343 CHARACTERISTICS OFTHE WALUS 00 ea masia erie 44 3 3 1 Geometry and construction SEQUENCE cessccccccscessessscssecscessssnesecsseesesssnssecssessasesesecsseeseasessseceeesaaaeseseceees 44 3 3 2 Material and model characteristics ccccccsssccesssncecssnececesnecesseececesseceseaaesesaeeecsesuecesseaaeeseeeeeseesaeeeeeegs 45 34 STRUCTURAL RESULTS reee een ae enea one Eon EEA Ee EEA Se E EEA INET CAER dana ten dddune adas dentista ti deeirami geni 47 3 5 SECTIONAL RESULTS ciat senat ciat a ide coed 50 3 677 DISCUSSION EEA GE i E AE cil 52 eTA GONGEUSIONS R E E A A A E E E ESE 54 3 8 ACKNOWLEDGEMENTS neisieiieiiiiiiiireriririiiiini iii iiie iiin i E rE iie EEE iE E EA e iein e ae 55 CHAPTER 4 PARAMETRIC STUDY OF CONSTRUCTION PROCESSES o oooccccccccccccooooocccccccccnnoncnnnccnnnccconnnnccccccnnnnnnnos 57 A T INTRODUCTION co coit cone comt ds ucossms ostenticos mames 58 4 2 METHODOLOGY immi dii been dates A a to 59 4 2 1 Model description tim i
40. application of the epoxy bonding agent The pull off test showed a wide range of results which arose from the experimental errors caused due the difficulties to perform the test on the irregular surfaces of both layers Structural behaviour of the bi layer diaphragm walls The experimental results of readings from inclinometers embedded in the walls were reproduced with a FEM numerical model running on the PLAXIS program A comparison of the results from the calibrated model and from the experimental campaign demonstrated a very good correlation which validated the model The FEM model considered cross sectional changes in the stages after spraying the second layer and reproduced both the qualitative and the quantitative displacements of the instrumented walls with a high degree of accuracy The structural behaviour of bi layer walls and mono layer walls with the same first layer SS thickness is similar in relation to their total envelope of bending moments and deformations This is because the increase in bending moments in the sprayed sections CS caused by the subsequent excavations are small compared with the change in moments that took place in the section not yet excavated SS Therefore the first layer SS governs the general behaviour of the walls The envelope of moments of the simple cross section i e the moment that have to be resisted by the SS cross section are smaller in the bi layer than in the mono layer alternatives 17
41. bi layer and mono layer walls due to their different stiffnesses The different deflections are recorded at the top of the wall because it is a free end in the structural configuration The maximum displacements difference in this case is 0 68 mm However the difference in the maximum displacements depth 6 0 m is minimal 0 02 mm in this case Meanwhile large differences in the bending moments can be seen in the TD case both above and below the excavation depth after each excavation stage see Fig 37c Regarding the displacements of this case they can be observed mainly between depths 6 0 m and 12 0 m as depths of between 0 0 m and 6 0 m are Luis Segura Castillo Parametric study of construction processes 73 strongly fixed by the upper slab depth 0 0 m and the intermediate slab depth 6 0 m Therefore despite being a quite stiff structural configuration a maximum displacement difference of the same order of magnitude 0 45 mm may be seen 4 3 5 Sectional results As an example reinforcement design of different walls is shown in Fig 39 The same representation used for Fig 31 is followed here It can be seen that the design inequalities M gt gt M A and M F gt M z see Segura Castillo Aguado de la Fuente et al 2013 are satisfied in each instance 104 pa ng 12 Depth m 14 16 ie A 18 20 1 20 400 200 0 200 400 400 200 0 200 400 600 400 200 0
42. compared to the MP series The only difference in the preparation of these series was surface wetting The results indicate that total saturation of the surface at the time of concreting reduces the bonding capacity between the two concretes 5 3 2 4 Shear stress on epoxy bonded surfaces Fig 49 shows a graph of the average values and represents the standard deviation at each age for the EP series Only one core could be tested at the first age t 3 days therefore the dispersion for this time cannot be assessed The variation differed greatly for the two other ages under consideration The widest dispersion 0 68 MPa was recorded at the second age t 6 days and a much smaller standard deviation 0 11 MPa was recorded at the third age t 35 days The wider variability of the results for the EP series may be attributed to difficulties in the application of the bonding agent that is inherent to the work After application of the product one hour must elapse before proceeding to place the second phase concrete Likewise the product must be applied within a period of approximately two hours after that period it crystallizes notably reducing the bonding capacity between concretes If we take into account the difficulty of accurately predicting concreting times when spraying it may be concluded that areas of varying strength in the bond will probably be found On the other hand since the product must be applied manually its collocation depen
43. de conglomerante org nico resinas al hormig n base Abu Tair et al 1996 Sin embargo no hay acuerdo entre investigadores sobre la idoneidad en materiales de conglomerantes hidr ulicos base cemento Momayez et al 2005 Los m todos m s difundidos para el ensayo a corte est n dise ados para ensayar probetas elaboradas en laboratorio Como ensayos apropiados para aplicar en testigos obtenidos en campo podemos mencionar el ensayo LCB Mir Recasens et al 2005 y el ensayo a corte directo guillotina Delatte et al 2000 Como ensayo para realizar in situ se puede mencionar el ensayo pull off Simon Austin Robins amp Pan 1995 Mediante el mismo se es capaz de evaluar la adherencia de la uni n sometida a esfuerzos de tracci n En particular J lio et al 2005 examinaron la correlaci n entre la adherencia a corte medida con el ensayo a corte oblicuo y la adherencia a tracci n obtenida con el ensayo pull off registrando una correlaci n linear A1 2 OBJETIVOS Los objetivos buscados en la realizaci n de este trabajo son dos e Estudiar la evoluci n a tempranas edades 2 6 y 35 d as de la resistencia del v nculo alcanzada entre el hormig n proyectado sobre el hormig n base previamente fresado que conforman el muro Bi Capa e Buscar un ensayo pr ctico para caracterizar la adherencia entre capas factible de ser utilizado como control rutinario en la implementaci n de esta tipolog a de elementos
44. desde el primer al ltimo d a a dar una mano y otro a Ricardo con quien la lucha se traslad tambi n fuera de la uni Agradezco al IET y la FING de la UdelaR que me han dado la oportunidad de hacer mis primeras armas en esta carrera en especial a Gemma Berardi el Guti y Atilio que me han incentivado y apoyado desde el comienzo y hasta el final de esta aventura que ahora culmina Al equipo de PERMASTOP en especial a Ra l Suarez con quien compartimos d as de lucha a pie de obra A todos mis compa eros de piso Gracias Bea Andrea gran compa ero de lucha hoy perdido en la batalla David Jada y Maribel Pero principalmente a Elena y Mar que me recibieron cuidaron animaron mimaron e hicieron posible un comienzo genial de esta aventura Hicieron que a mi llegada a esta nueva vida me sintiera como en casa gracias a ustedes Arag fue mi hogar hasta que lleg Coralinda a poner orden aunque en estos a os hemos podido descubrir con alegr a que debajo de esa Coraza hay un Coraz n Mucha gente grande en Barcelona Oriol y Mireia que tambi n intentaron hacerme bailar Tommy Lea Maco me hicieron sentir m s cerco del barrio Roberto y Alba y sus verduras org nicas Susana siempre en el aire A Gerard el orco de la monta a Maria Isabel la presidenta Alejo Martha Alejandra te visitar en Celaya Guanajuato M jico y Jorge un gran grupo Mat as el Borges chileno nuestros proyectos reci n comienza
45. due to their associated problems In most cases the bond surface in a direct shear test is in fact subjected to shear stress and to slight bending In some tests shear stress is combined with a normal either tensile or compressive stress Saucier Saucier et al 1991 devised a test for assessing shear bond strength under different compression levels and Austin Simon Austin et al 1999 considered bond failure as an envelope covering all possible normal direct stress states However when stress was introduced into the shear plane by means of steel plates it caused stress concentrations at the edge of the bond plane Smaller stress concentrations reduced the scatter in the test results Momayez et al 2005 The most widespread methods for the shear test are designed to test samples produced in a laboratory The LCB test Mir Recasens Martinez amp P rez Jim nez 2005 and the Guillotine Direct Shear Test Delatte et al 2000 are worth mentioning as suitable tests for cores obtained in the field The aim of this research is to study the evolution of bond strength at early ages 2 6 and 35 days that is obtained between sprayed concrete and a previously milled base layer of concrete which together make up the bi layer diaphragm wall The study also analyzes the influence of contact conditions for which purpose several other alternatives have been added such as priming with an epoxy resin coat and water saturation of the contact s
46. favourable to the bi layer systems Finally an increase in the total amount of steel 2 0 is still registered 3 7 CONCLUSIONS A design method for the bi layer diaphragm walls a new type of slurry wall has been presented It allows two levels of comparison the first of which is based on the structural analysis and the second on the final design where the comparison includes final material use The structural behaviour of a conventional ML wall 60 cm width RC layer has been compared with two BL alternatives 60 cm and 55 cm width RC layer plus 10 cm width SFRC layer Furthermore starting with these wall types several systems to deal with leakages have been added to the comparison The main conclusions are summarized in the following points There exists a potential of reduction in the reinforcement of the RC layer of the diaphragm walls through the structural use of the SFRC layer This potential is measured by the area of moments envelope covered by the simple section M This area is reduced 21 and 26 in both BL alternatives compared with the ML wall However it is not possible to take advantage of all this potential in the design process for two reasons that are explained as follows The increase from the MS to the M Ea is on average 123 of the MS at depths of between 2 5m and 11 0m Besides the increase from the M gt to the M S are in this case between 15 and 20 of the M gt This means that if the SS section i
47. is considered for the assessment of the shear strength 50 tF35 days 45 A y gt e D 6 days 5 40 tF2 days O v Q o 3 o 2 a 35 43 ist phase 8 I A A 2nd phase 30 0 80 100 120 140 160 1 80 Shear strength t MPa Fig 54 Influence of the strength of base concrete on shear strength MP case According to the results obtained for t 6 days and t 35 days the concrete with a lower strength first phase remained almost unchanged however the higher strength concrete second phase showed a considerable increase in strength and at the same time the bond shear strength increased Additional research is necessary but the experimental data and results suggest as advanced by Julio that the formula to assess shear strength may be improved by incorporating higher concrete strengths 5 4 CONCLUSIONS This experimental study has studied bi layer diaphragm concrete walls and more particularly the bond between the second phase of sprayed concrete placed over a first phase of surface milled concrete The following conclusions have been drawn e The test proposed for determining shear strength between concretes poured in different phases yielded valid results With milled surfaces the coefficient of variation CV was lower than 15 e Shear strength on the milled surface increased with the age and the strength of the second phase concrete in line
48. seen that the reduction is generally greater in cases with a 55 cm first layer thickness However even though similar displacements are obtained for each final structure configuration and construction sequence in all cases the incorporation of the second layer is not enough to compensate for the extra displacements that take place if the cross section of the first layer is reduced from 60 cm to 55 cm 4 3 1 Influence of the number of spraying stages There are construction reasons to analyze the number of spraying stages On the one hand spraying the concrete for the second layer in a single stage may require sophisticated equipment e g a spraying robot and might even be impossible in large scale sections On the other hand spraying is easier if divided into sections of a few meters when sprayed from various levels as the excavation progresses and it is even possible to do so with manually operated equipment Likewise spraying in several stages makes it easier to coordinate this task with the excavation and the installation of props thereby reducing construction time However concrete joints between the different spraying stages where the strength of the concrete may be weaker and its watertightness less effective appear in this case in the SFRC layer This parameter figures in the structural calculations as indicated below and influences the development of bending moments which may influence the structural contribution of the SFRC layer
49. that debonded at the time of their extraction 5 2 2 Shear test The modified LCB test was chosen for the shear test from among those described in the introduction Fig 43 shows a sketch of the device and a photograph of the test This test was chosen for two reasons on the one hand it meant that a shear test could be performed on the extracted cores and on the other hand it made it easier to test irregular bonds between layers a problem reported in the case of the guillotine test Delatte Jr et al 2000 as it leaves a small space between the load introduction edges However due to this separation the bending component acting on the bond to be tested was higher The test is based on standard NLT 328 08 NLT 382 08 2008 intended for the assessment of bonding in pavement layers made of bituminous materials Mir Recasens et al 2005 which are much more ductile materials than concrete with much greater sensitivity to temperature variations Because of this the temperature control chamber was removed and the displacement speed of the loading piston was reduced from 2 5 mm min to a value in the order of those used in shear tests on concrete 0 25 mm min Mirsayah amp Banthia 2002 Ray Davalos amp Luo 2005 Wall amp Shrive 1988 A thin neoprene sheet was placed Bi layer diaphragm walls Experimental and numerical analysis 64 Chapter 5 between the device and the core in order to reduce stress concentration
50. the SFRC layer Therefore the curvature increase of the bi layer wall is smaller than in the mono layer alternative as the bending moment increases In this instant as the wall is more restrained in the lower part at depths of below 5 0 m owing to the embedded end of the wall and the stiffer lower anchors the upper free end of the wall is dragged outwards at greater extent The envelope of flexural design moments M obtained for all three wall types for both the SS and for the CS sections are shown in Fig 24a A partial factor of y 1 5 was applied to the actions of the ground on the wall The positive moments of the envelope are the same for both kinds of sections In turn while there is a single envelope for the negative moments of the mono layer wall the envelope is broken down into two envelopes for the bi layer walls The envelopes of maximum moment until the CS was completed i e until the SFRC was sprayed are shown by a continuous line MS and the envelopes where the CS was completed by a dashed line Mm The way these envelopes were generated is explained below in greater detail In general terms the envelopes are qualitatively similar and within the order of magnitude of the reference case Carrubba amp Colonna 2000 detailed 1 in Fig 25 Bi layer b s wd EL hm o w EB a y 4838 8888 Depth m Depth m 8 4310 3415 3172 8 Bending Moments Envelope area RN m m 8 8
51. the excavation process the wall was supported by up to 4 rows of ground anchors vertically spaced at 3 0 m and horizontally spaced at 5 0 m in the two upper rows Superior Anchorages S A and at 2 5 m in the two lower rows Inferior Anchorages I A 0 00 m b Simple Section SS Vertical Ground reinforcement 1St layer conventional RC c Compound Section CS Bond d A 204 layer plane sprayed SFRC Fig 21 a Model geometry Anchorages and slabs positions b Simple Section c Compound Section The bottom up construction sequence of the three alternatives is detailed in Table 12 The stages are divided in 5 groups the first 4 of which correspond to the Excavation Exc works required for each of the 4 Luis Segura Castillo Structural and sectional analysis 45 Anchorage Anc installations In the fifth group apart from a small final excavation the slabs are built and the 4 anchorages removed Anc Out Table 12 Construction stages sequence Group Depth m MLso BL 604 10 and BLss 10 Wall constr Wall constr 1 1 75 Exc 1 Exc 1 1 50 Anc l Anc l 3 15 Exc 2a 3 50 Spray 2a 2 4 75 Exc 2 Exc 2b 4 50 Spray 2b 4 50 Anc 2 Anc 2 6 75 Exc 3a 6 50 Spray 3a 3 7 15 Exc 3 Exc 3b 7 50 Spray 3b 7 50 Anc 3 Anc 3 9 75 Exc 4a 9 50 Spray 4a 4 10 75 Exc 4 Exc 4b 10 50 Spray 4b 10 50 Anc 4 Anc 4 12 50 Exc 5 Exc 5 12 25 Spray 5 s
52. the method when compared with equivalent mono layer wall alternatives O5 Study the influence of the different constructions processes related to this type of walls 06 Evaluation of the profitability through therorethical comparisons of different bi layer and mono layer walls M8 Parametric study based on the using the overall design method M9 Disseminate the results O7 Publish in high level refereed journal papers and present papers at relevant conferences M10 It can be seen that the thesis consists of four journal papers two of them already published P 1 and P 4 one accepted for publication P 2 and the last one already submitted and currently under the 2 review after the Journal having asked for some changes P 3 all of them in renowned international journals and one conference papers already published C P 1 The PhD candidate was the first author of all the papers The papers planning state of the art methodology analysis and conclusions were performed entirely by the PhD candidate with the recommendations of his advisors The writing of almost the totality of the four papers was also done by the candidate Furthermore all the papers were written during the doctoral studies period The thesis is structured around the presented papers Each of the journal papers conform a self contained chapter of the main body of the thesis as is schematically shown in Fig 5 In the figure the common part
53. the same figure A fine global coarseness was taken for the general mesh automatically defined by the program and refined in the vicinity of the bottom of the plate representing the diaphragm wall A similar domain was used for Wall W45 differentiated mainly by the anchorage distribution The Mohr Coulomb elastoplastic model assessed all four kinds of soils It is considered sufficient for the analysis as the study focuses more on the structural behaviour of the bi layer walls than on soil behaviour and in view of the practical uncertainties involved in the definition of advanced model parameters for soil behaviour The selected elements were 15 node triangular finite elements under plane strain a oe SEES Fig 9 Finite Element mesh of W35 wall The soil parameters determined by the geotechnical study and used in the model are shown in Table 7 total unit weight y cohesion c friction angle coefficient of lateral earth pressure at rest Ko estimated by the expression of Jaky Terzaghi Peck amp Mesri 1996 initial depth of the level zni strength reduction factor for soil structure interface Rimer Poisson coefficient v the Young s modulus approximated from Bi layer diaphragm walls Experimental and numerical analysis 28 Chapter 2 geotechnical parameters Ep and the one determined in back analysis Era As the phreatic level is below the maximum excavation no water flows are consi
54. thesis is centred in the structural behaviour of the diaphragm walls Furthermore when possible cases with a large security factor against these modes were used to avoid any possible interaction with the structural failure Luis Segura Castillo Introduction 7 There are currently in use a variety of methods to check the previously mentioned ultimate states in order to design embedded walls A State of the art about the different available calculation methods was performed by Delattre 2001 were the evolution and development of the different methods is described According to Delattre 2001 the diversity of methods arises from the complexity of the geotechnical structure which is both supported but also loaded by the soil Delattre 2001 classified the methods in the following categories 1 classical methods 2 subgrade reaction method 3 finite element method and 4 empirical methods The different methods are able to tackle with different precision the different ultimate states that should be checked The classical design methods are based on the classical Coulomb and Rankine soil behaviour methods and their extensions These methods are centred on the pressures exerted on the structures by the soil not taking into account the deformations of the structures and the adjacent soil The subgrade reaction method based on the Winkler model was later developed It is able to take into account the soil and structure deformations con
55. three groups 1 Thin sheet elements As conventional reinforcement cannot be used the fibres are used as the primary reinforcement 2 Elements subjected to heavy locally applied loads or deformations e g tunnel linings blast resistant structures or precast piles 3 Elements where cracking due to humidity or temperature variations must be controlled e g slabs and pavements In this case fibres are often referred to as secondary reinforcement As an extra advantage the placing of the mesh or reinforcing bars is eliminated in application where FRC is used It is also worth mentioning that in these applications the fibre reinforcement is not essential for the structural safety Zollo 1997 A more recently important milestone for the FRC as a structural material in Europe happened with the publication of design codes and recommendations which provided a scientifically founded consistent and coherent framework for the design of FRC elements di Prisco Plizzari amp Vandewalle 2009 In this sense it is worthwhile mentioning in order of time of publication the German guidelines DBV 2001 the RILEM TC 162 TDF recommendations RILEM TC 162 TDF 2003 the Italian guidelines CNR DT 204 CNR 2006 the Spanish code EHE 08 CPH 2008 and the Model Code 2010 FIB 2010 An exhaustive analysis and comparison of the mentioned codes was performed by Blanco Pujadas de la Fuente Cavalaro amp Aguado 2013 Besides these applic
56. used in the structural study The soil was modelled with the Hardening Soil model HS Schanz et al 1999 and the wall and supports with linear elastic elements In the FEM model the stiffnesses were updated from the simple cross section to the compound cross section in the corresponding wall sections that had been sprayed after each of the spraying stages No movements were considered during struts and slabs installation and the walls were considered wished in place i e the stress changes or displacement of the wall installation in the soil are not considered in the model Bryson amp Zapata Medina 2012 In all cases diaphragm walls of 20 m in height were built for subsequent excavation work to a depth of 12 5 m and with embedded footings of 7 5 m in depth No adjacent buildings were considered i e no external loads were introduced in the models A sandy soil Lake sand layer and its parameters were taken from a case presented elsewhere Hashash et al 2010 This is a good quality only slightly deformable soil The type and characteristics of the finite elements the mesh discretization and its boundary conditions as well as properties taken for the wall anchorages and slabs are the same as those in Segura Castillo Aguado de la Fuente et al 2013 The struts were modelled with fixed end anchors A normal stiffness of EA 2 00 10 kN m m for the superior strut and of EA 4 00 10 kN m m for the inferior one was selecte
57. walls was constituted by a conventional reinforced concrete diaphragm wall with a compression strength at 28 days of f 30 MPa UNE EN 12390 3 UNE EN 12390 3 2003 the mixture composition is given in Table 19 For the characterization of the concrete samples were taken at the time of concreting the walls with which the compressive strength of the first phase concrete was determined UNE EN 12390 3 UNE EN 12390 3 2003 Limestone type aggregates were used in both phases Table 19 Concrete dosages Component First phase Second phase concrete concrete Cement II A M 42 5 R KG m3 365 Cement 1 52 5 R KG m3 450 Corrector sand 0 2 KG m3 355 Sand 0 4 KG m3 970 975 Aggregate 4 12 KG m3 200 300 Gravel 12 20 KG m3 630 Plasticizer cement in kg 0 8 Nano silica cement in kg 1 5 Superplasticizer cement in kg 1 5 Water cement ratio 0 47 0 40 Metallic fibres kg m3 30 Once the excavation was finished cold milling of the exposed wall took place in order to even out and prepare the surface and to increase its roughness so as to improve the bond of the sprayed concrete layer Milling was performed with a concrete miller attached to the end of a backhoe Fig 42a A negative aspect of this type of mechanical treatment is that it can give rise to micro cracks which weaken the surface J lio et al 2004 Talbot et al 1994 Fig 42b shows a photo of the surface finish highlighting a specific area Subseque
58. where the wall is directly supported by the finished slabs NU is the number of underground levels of the finished structure configuration where two alternatives are considered 2 and 4 underground levels represented by 2u and 4u 6 0 m and 3 0 m high respectively NS is the number of spraying stages performed to cover the whole external wall surface where four alternatives are considered 0 1 2 and 4 spraying stages represented by OS ML wall 1S 2S and 4S respectively DS stands for the depth of the excavation at the moment where the last spraying stage was carried out Four alternatives are considered M where there is no spraying ML wall A the depth of the final excavation i e 12 5 m B one meter before the final excavation depth i e 11 5 m and C two meters before the final excavation depth i e 10 5 m As an example a Wall labelled BL 55 10 BUs 4u 2S A corresponds to a bi layer wall with a RC layer thickness of 55 cm and a SFRC layer thickness of 10 cm built with the Bottom Up construction sequence Luis Segura Castillo Parametric study of construction processes 61 using struts with a final structure of 4 underground levels and a two stage spraying process the second of which is sprayed when an excavation depth of 12 5 m is reached Table 16 Parameters and alternatives for each case Parameter Alternatives Description Number of layers ML Mono layer NL BL Bi layer I
59. with the maturity formulas e Shear strength when assessed with this test was not dependent on the direction of the milling on the bond surface Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 93 e Bonding capacity after spraying decreased on the surfaces that had previously been saturated with water e Finally the results for in situ preparation of the surface with epoxy showed great variation with a CV of up to 57 in the worst case In view of these results it is not advisable to use epoxy products on bi layer walls or large surfaces in general unless the application and execution times of the bonding agent and the concrete spraying are rigorously controlled Improvements to the applied test have been proposed so that it may potentially become a valid test for samples produced in the laboratory and for cores extracted on site Future research work will focus on a variations in the angle of the failure plane and b the production of notches in the cores and the way these may influence bond strength 5 5 ACKNOWLEDGMENTS The authors would like to thank PERMASTOP TECHNOLOGIES for financial support CTT 8062 physical resources and the assistance of their staff especially Ra l Suarez and Tomas Duran without whom this work could never have been completed and the Instituto de Estructuras y Transporte of the Universidad de la Republica Uruguay for financial support Funding was also made avai
60. 0 03 BL 60 10 BUs 4u 2S B 4004 549 265 9 303 2 35 8 13 5 9 73 0 01 BL 60 10 BUs 4u 2S C 4052 529 309 8 347 6 49 2 0 0 9 70 0 03 ML 55 0 BUs 4u 0S M 4089 267 6 55 7 10 23 BL 55 10 BUs 4u 1S A 3909 197 264 0 299 0 55 8 0 1 10 22 0 01 BL 55 10 BUs 4u 2S A 3797 294 267 8 302 8 54 0 1 7 10 24 0 00 BL 55 10 BUs 4u 4S A 3643 475 270 2 305 3 55 8 0 1 10 17 0 06 BL 55 10 BUs 4u 2S B 3648 460 234 1 268 7 42 1 13 6 10 24 0 01 BL 55 10 BUs 4u 2S C 3695 413 268 7 303 8 55 3 0 4 10 21 0 02 ML 60 0 B Us 2u 0S M 4874 345 1 66 5 10 25 BL 60 10 BUs 2u 2S A 4202 719 314 8 352 7 53 3 13 1 10 12 0 13 BL 60 10 BUs 2u 2S B 4026 914 319 3 357 2 54 3 12 1 10 11 0 14 ML 55 0 BUs 2u 0S M 4436 307 7 73 6 10 88 BL 55 10 BUs 2u 2S A 3815 657 279 1 314 3 62 2 11 4 10 69 0 19 BL 55 10 BUs 2u 2S B 3667 824 283 0 318 2 63 3 10 3 10 69 0 19 ML 60 0 BUa 2u 0S M 6564 488 0 188 1 25 28 BL 60 10 BUa 2u 2S A 5685 1033 468 8 508 4 174 5 13 5 25 29 0 01 BL 60 10 BUa 2u 2S B 5226 1566 479 4 519 1 176 8 11 2 25 26 0 02 ML 55 0 BUa 2u 0S M 6104 454 9 211 6 27 16 BL 55 10 BUa 2u 2S A 5277 954 435 5 472 1 195 7 15 9 26 95 0 21 BL 55 10 BUa 2u 2S B 4850 1458 447 6 484 9 204 3 7 3 26 73 0 43 ML 60 0 TD 2u 0S M 4987 355 8 69 7 7 29 BL 60 10 TD 2u 2S A 4598 432 328 1 366 1 59 6 10 2 6 84 0 45 BL 60 10 TD 2u 2S B 4451 623 298 5 336 2 48 1 21 7 6 84 0 45 ML 55 0 TD 2u 0S
61. 04b According to these hypotheses the ultimate bending moment My is calculated and compared with the maximum design bending moment Ma calculated by the structural analysis for the most unfavourable construction stage and for each kind of section The compressive behaviour of the concrete see Fig 20b was simulated on the one hand by considering the constitutive law proposed in EC 2 EN 2004b On the other hand the tensile response of the SFRC was simulated through constitutive law o e as suggested in RILEM TC 162 TDF 2003 Finally the mechanical performance of the steel bars was simulated with the bilinear diagram presented in Fig 20c CS Cross Baa Section HG este I T 03 Ec ESFRGb 01 Esp Strain profile Je Cb Es po c Ence aN A Stress profile M Fig 20 a Sectional discretization b SFRC and c steel bar constitutive equations In addition to the internal equilibrium conditions the following hypotheses are also considered 1 the sections remain plane before the application of the external forces or after imposing fixed strains 2 failure of the composite section is achieved when there is either excessive compressive strain in the upper concrete layer ERc4 3 5 0 and or excessive elongational strain in the tensioned steel bars s 10 0 c and 3 a perfect bond between the concrete and the rebars as well as between the RC and the SFRC layers Regarding Bi lay
62. 08 1 2 THESIS OBJECTIVES The aim of the research was to develop a new type of slurry wall The bi layer diaphragm wall The method to construct it is based on the tanking solution described above where a second waterproof layer is casted against the diaphragm walls The bi layer diaphragm wall is made of two bonded concrete layers poured and then sprayed in separate stages The first is a conventional Reinforced Concrete RC diaphragm wall Once this wall attains the necessary strength soil within the perimeter is excavated and removed and the second layer this time of sprayed Steel Fibre Reinforced Concrete SFRC with a waterproof additive is applied The solution is schematically presented in Fig 2 Compound Section pe eS td ea es Bond plane Area with i sprayed be concrete Area as yet unsprayed rn 2nd layer sprayed SFRC Ground n Simple Sectio Vertical reinforcement lt 4st layer conventional RC Fig 2 Bi layer diaphragm walls a general scheme b compound section c simple section and d spraying of an experimental wall The idea is to maximize the functional attributes of the second layer allowing it to play a structural role in addition to its initial intended purpose waterproofing Due to the structural role of the second layer the thickness and reinforcement of the first layer may be reduced becoming an attractive structural solution if the wate
63. 122 7 16 4 16 47 0 23 In the Loose Sand case it can be seen that both the bending moments and the displacements have increased considerably compared with the base case BL 60 10 BUs 2u 2S A in agreement with the change made However a reduction of 16 4 kg in the amount of steel within the range of the other cases was still possible 4 5 CONCLUSIONS By means of a numerical study a parametric analysis of six relevant bi layer diaphragm wall parameters has been performed Several construction impacts have been detailed through the paper and the steel reinforcement and displacements reduction have been quantified The main conclusions are 4 6 In all cases the main flexural strength is provided by the strength of the first layer with conventional bar reinforcements and a secondary flexural contribution by the second layer with SFRC This contribution increases the ultimate strength of the simple cross section by between 8 and 15 Feasibly all final structural configurations and construction sequences could reduce the steel reinforcement of the RC layer by taking full advantage of the strength added by the SFRC layer Using the appropriate spraying sequence a reduction in steel reinforcements of over 10 kg can be obtained in every configuration and sequence reaching a maximum reduction of up to 21 7 kg This represents a percentage reduction of 7 0 of the total bending reinforcement and of 45 1 over the extra b
64. 13 Bi layer diaphragm walls Experimental and numerical structural analysis Engineering Structures 56 154 164 doi 10 1016 j engstruct 2013 04 018 Segura Castillo L amp Aguado de Cea A 2012a Bi layer diaphragm walls Evolution of concrete to concrete bond strength at early ages Construction and Building Materials 31 1 29 37 doi 10 1016 j conbuildmat 2011 12 090 Segura Castillo L amp Aguado de Cea A 2012b Bi layer diaphragm walls Early ages concrete to concrete bond strength assessed through shear and pull off tests in spanish In XXXV Jornadas Sudamericanas de Ingenieria Estructural Rio de Janeiro ASAEE Segura Castillo L Josa A amp Aguado A n d Bi layer diaphragm walls Parametric study of construction processes Engineering Structures Submitted Sherif A S amp Kudsi T N 1975 Reliability of underground concrete structures under water ingress attack In Fourth International Symposium on Uncertainty Modeling and Analysis 2003 ISUMA 2003 Vol 52 pp 40 44 IEEE doi 10 1109 ISUMA 2003 1236138 Shohet I M amp Galil I 2005 Decision Support System for Waterproofing of Below Grade Structures Computer Aided Civil and Infrastructure Engineering 20 3 206 220 doi 10 1111 1467 8667 2005 00388 x Talbot C Pigeon M Beaupr D amp Morgan D 1994 Influence of surface preparation on long term bonding of shotcrete ACI Materials Journal 91 6 560 566 Retrie
65. 4 ETJ exc225 ano a e 6 0 i 60 4 70 4 I 70 4 8 0 4 I 8 0 4 exc630 elim200 9 0 i 9 0 4 10 0 10 0 5 0 4 0 30 20 10 00 1 0 100 80 60 40 20 O 20 40 60 Displacements mm Bending moment kN m Fig 15 Horizontal displacements and bending moments obtained by the adjusted PLAXIS model for Wall W45 2 6 DISCUSSION DESIGN PROCESS AND COMPARISON A bi layer wall case study namely Wall T BLW35 with a cast layer of 35 cm and a sprayed layer of 10 cm is discussed in this section Its design method takes advantage of the compound section A comparison with two conventional single layer walls quantified differences with regard to the performance of bi layer walls The first wall T W35 had the initial thickness of the bi layer wall 35 cm and the second T W45 had the final thickness 45 cm All cases were implemented with the previously described model The analyses Luis Segura Castillo Experimental and numerical structural analysis 35 presented below cover the structural level leaving the sectional analysis for a further paper which will apply a similar strategy to that presented in de la Fuente Aguado de Cea et al 2012 2 6 1 Description of case studies All the general characteristics of the model of Wall W45 described in section 2 3 are used in these models except for the modifications specifically described below The construction stages are indicated in Table 10 Th
66. 5 1 0 15 2 0 23 22 21 20 8 5 duyn huone Depth m 30 25 20 15 10 5 0 Horizontal Displacements mm Fig 38 Displacements for different construction sequences Large differences can be seen in the displacements of the three construction sequences The smallest displacements were registered for the TD case followed by the BUs and finally the BUa cases This order reflects the support stiffness Moreover the anchorages have extra flexibility provided by the deformability of the ground surrounding the bulb area besides the flexibility of the element itself The displacements although small fall within the range of expected values in the Moormann database Moormann 2004 They reflect reasonable values if we remember that a relatively stiff soil was used in this study and that the supports were placed early in the excavation process before appreciable displacements were recorded The figure shows that once again the influence of the bi layer dark lines in the displacements is small compared with the differences in the construction sequence Nonetheless the differences between bi layer and mono layer in the TD and BUa are greater than in the BUs cases already described in section Error Reference source not found In the BUa case see Fig 37b these large differences arise as larger moments are recorded after the first spraying which subsequently causes a greater difference in the curvatures of the
67. 5 PE Tl 2 Rech RI 9 2 41 51 GC 35 t1 35 PE Tl 2 Rech R S 264 0 20 13 2 GA 45 t1 45 PF Tl 2 Rech R S 196 0 20 9 8 GC 45 t1 45 PF T1 2 Acept R I 16 0 303 9 41 51 7 32 GB 45 T1 45 PF T1 2 Acept R I 12 9 168 9 41 51 4 07 GE 45 t1 45 PF Tl 2 Rech R S 22 2 20 1 11 GD 45 t1 45 PF T1 2 Acept R I 8 5 73 1 41 51 1 76 GA 35 t2 35 PE T2 6 Acept R I 11 0 313 0 41 51 7 54 GB 35 t2 35 PE T2 6 Acept R I 9 5 244 1 41 51 5 88 GC 35 t2 35 PE T2 6 Acept R I 8 5 34 0 41 51 0 82 GD 35 t2 35 PE T2 6 Acept R I 12 6 70 2 41 51 1 69 GE 35 t2 35 PE T2 6 Acept R I 11 2 114 2 41 51 2 75 GA 45 t2 45 PF T2 6 Acept R I 16 0 144 0 41 51 3 47 GB 45 t2 45 PF T2 6 Acept R I 15 7 302 2 41 51 7 28 GC 45 t2 45 PF T2 6 Acept R I 14 3 190 1 41 51 4 58 GD 45 t2 45 PF T2 6 Rech R S 180 0 20 9 GE 45 t2 45 PF T2 6 Rech R S 186 0 20 9 3 GC 35 t3 35 PE T3 35 Acept R I 10 7 94 9 32 17 2 95 GD 35 t3 35 PE T3 35 Acept R I 13 0 84 9 32 17 2 64 GE 35 t3 35 PE T3 35 Rech R A 233 0 20 11 65 GB 35 t3 35 PE T3 35 Rech R A 223 0 20 11 15 GA 35 t3 35 PE T3 35 Rech R A 263 0 20 13 15 GC 45 t3 45 PF T3 35 Acept R I 14 4 99 1 32 17 3 08 GD 45 t3 45 PF T3 35 Rech R A 58 0 20 2 9 GA 45 t3 45 PF T3 35 Rech R A 167 0 20 8 35 GB 45 t3 45 PF T3 35 Rech R A 157 0 20 7 85 GE 45 t3 45 PF T3 35 Rech R A 85 0 20 4 25 C digos de tipos de Rotura R S Rotura interna superficial del hormig n proyectado R I Rotura por la interfase entre el hormig n pr
68. An example of bending moments obtained for a generic bi layer case is shown in Fig 31 The design bending moments M4 are represented with blue lines The envelopes both positive and negative are broken s down into two in the bi layer cases The maximum moments envelope of the simple cross section Ma until the stage where the compound cross section is completed i e when the SFRC is sprayed are plotted with a continuous line The bending envelopes developed when the compound cross section had already been completed M E are plotted by a dashed line The M that are shown include a partial security factor y 1 5 The way these envelopes are generated is explained in Segura Castillo Aguado de la Fuente et ss al 2013 in more detail The figure also shows the ultimate bending moments My M U with a continuous G black line and My with a dashed black line MBA min 0 MB As min Aco ME Ag mintAs exd H MB As m 6 Bi layer area work 4 8 sprayed section 10 i 12 Depth m 14 mss 16 MS 18 20 400 200 0 200 400 Bending Moments kKN m Fig 31 Example of bi layer wall bending moment envelopes The design moments increase between M hi and M4 can be resisted by the ultimate moment of the cross section when the SFRC layer is added M o Therefore the area between the M and the M T envelopes indicated in the figure with a pattern fill which will be called
69. Bi layer diaphragm vvalls Experimental and numerical analysis Doctoral thesis by Luis Segura Castillo Directed by Antonio Aguado de Cea Alejandro Josa Barcelona September 2013 Universitat Polit cnica de Catalunya Departament d Enginyeria de la Construcci DOCTORALTHESIS UNIVERSITAT POLIT CNICA DE CATALUNYA BARCELONATECH Escola de Doctorat re az Curs acad mic Acta de qualificaci de tesi doctoral Curs acad mic Nom i cognoms Programa de doctorat Unitat estructural responsable del programa Resoluci del Tribunal Reunit el Tribunal designat a l efecte el doctorand la doctoranda exposa el tema de la seva tesi doctoral titulada Acabada la lectura i despr s de donar resposta a les q estions formulades pels membres titulars del tribunal aquest atorga la qualificaci _ NO APTE _ APROVAT _ NOTABLE _ EXCEL LENT Nom cognoms i signatura Nom cognoms i signatura President a Secretari ria Nom cognoms i signatura Nom cognoms i signatura Nom cognoms i signatura Vocal Vocal Vocal i d de de El resultat de l escrutini dels vots emesos pels membres titulars del tribunal efectuat per l Escola de Doctorat a inst ncia de la Comissi de Doctorat de la UPC atorga la MENCI CUM LAUDE C s El No Nom cognoms i signatura Nom cognoms i signatura Presidenta de la Comissi de Doctorat Secret ria de la Comissi de Doctorat
70. EL The numerical model was calculated on the commercial geotechnical finite element software package PLAXIS Brinkgreve 2002 Experimentally determined parameters or those calculated from them were inputted into the model Then a final adjustment of the Young s modulus of the soil and the thicknesses of each layer were performed through a trial and error procedure These two parameters were chosen because of the well known difficulties in determining the soil elastic modulus Hsiung 2009 and because of uncertainties over wall thicknesses the thickness of the grabbing tool gave a minimum value for the cast layer and information from the extracted cores gave a range of values for the second layer Segura Castillo amp Aguado de Cea 2012b A 2D Finite Element Model FEM was used to represent a cross section of wall and soil The modelled panels were placed in the centre of the wall 9 40 m away from the nearest corner for Wall W35 and 6 63 m for Wall W45 Considering that this distance is in the same order as the excavation and additionally that the vertical joints between panels reduce horizontal stiffness it is reasonable to assume that the constraints caused by the boundary effects at the corners of the walls were negligible The domain used in the analyses for Wall W35 is shown in Fig 9 Horizontal fixity was assumed for the vertical boundaries and horizontal and vertical fixities were assumed for the bottom boundary as shown in
71. F C i y poh em GC 45 i y GE 45 T2 ay a 4573 GC 45 T3 1 f o GD 45 T2 o E I o GA 45 T30 D 4573 odescart t3 I iS BA 45 T2 I l GC 45 T2 o o GA 45 T1 l GB 45 T2 i i i Ga mo Ga i i i i extras Y para Leyenda PULL OFF testigo corte auxiliar testigo corte P35 A i P35 B l GC 35 T3 44 GD 35 T3 l I o o Fewer t desa baste de 35 Tab GA 35 T2 l i o GE 35 T2 GP 35 134 GA 35 T3 l I GC35T1 GD35T1 GE35fi GB35 TIGA 35 T1 r O om 1 os 0 i hd o EY E ARE ROR EE A QI ape 2 0m 4 2 0m y Fig 58 Esquema de posiciones de extracci n de testigos para el ensayo de corte y de realizaci n de ensayos pull off in situ para las distintas preparaciones e imagen de la pantalla luego de realizadas las extracciones A1 3 2 Zonas de extracci n de testigos Para poder relacionar los resultados de ambos ensayos se extrajeron en cada zona y para cada tiempo el testigo necesario para el ensayo de corte junto al lugar donde se realiz el ensayo pull off En la Fig 58 se pueden observar las posiciones en las que se realizaron las extracciones y los ensayos Si bien originalmente se eligi una distribuci n regular para realizar los ensayos ver la distribuci n de los ensayos T1 correspondientes a la primera edad debido que el paramento no era completamente plano sino que presentaba un relieve con suaves protuberancias generadas por el proyec
72. M 4666 317 3 77 8 8 10 BL 55 10 TD 2u 2S A 4289 423 291 4 326 7 69 3 8 5 7 55 0 55 BL 55 10 TD 2u 2S B 4160 598 273 3 308 4 61 0 16 8 7 56 0 55 Table 17 includes information on bending moment envelope areas as described below for both the simple SS 4 Oia cross section EA M d and the compound cross section EA M4 internal ultimate moment My for the negative extra reinforced cross section broken down both for the simple cross section M li and the compound cross section M z weight of extra reinforcement steel W A used in each case as defined in section 4 2 1 steel weight variation AVV A ex comparing each case with the corresponding mono layer alternative i e the first case in each group maximum displacement rmax and maximum displacement variation A Hmax Once again in comparison with the corresponding mono layer alternative Although something different is specifically indicated all the values in the paper correspond to a 1 m wide strip of wall Bi layer diaphragm walls Experimental and numerical analysis 64 Chapter 4 Only the vveight of the extra steel reinforcement is included in Table 17 as the vveight of the minimum reinforcement is constant for each 1 layer thickness i e 261 6 kg for W1 60 cm and 251 8 kg for W1 55 cm and the secondary reinforcement is considered to be similar in all cases The weights shown correspond to a 1 m wide wall
73. Structural level analysis b Sectional level analysis c Bonding between layers and d General design and optimization Of these the structural level behaviour was partially reported in Segura Castillo Aguado amp Josa 2013 and the bond analysis in Segura Castillo amp Aguado de Cea 2012a The main objective of this paper is to analyze the structural and sectional behaviour of the bi layer diaphragm walls The overall design method is presented With it the contribution of each layer is quantified placing special emphasis on the SFRC layer contribution 3 2 METHODOLOGY A study in the form of an uncoupled structural section analysis based on the hypothetical case of various bi layer diaphragm walls was performed to fulfil our objectives A 2D Finite Element Model FEM was selected to analyze the structural behaviour A numerical rather than a simplified model is necessary as the constructive sequence is considered Carrubba amp Colonna 2000 which includes the cross section changes that take place when the SFRC is sprayed together with general wall and soil properties The literature contains many studies that utilize these models to analyze ground movements caused by deep excavations due to their importance in the prediction of possible damage to adjacent buildings during excavation process e g Hsiung 2009 Khoiri amp Ou 2013 G T C Kung et al 2007 G T C Kung 2009 On the other hand fewer studies e g
74. able 14 Reinforcements and My of the different wall types Reinforcement Position My Wall Type qr 0 M 5 M S mm cm m m kKN m kN m Asmin 16 24 20 0 0 0 186 ML o As 12 16 17 5 12 0 337 As 10 22 11 5 5 0 263 Ag min 16 24 20 0 0 0 186 223 BL o 10 Agai 10 12 17 5 12 0 326 As 10 24 11 5 5 5 257 294 Agsmini 16 25 20 0 0 0 161 195 BLss 10 Aga 16 28 17 5 11 5 300 As 912 28 11 5 5 5 240 275 Anchorage length not included The ultimate resistance of the CS cross section where the positive moment reinforcements were placed was not calculated since this reinforcement is placed at depths lower than 12 5m where there is no second layer The ultimate moments obtained with the aforementioned reinforcements cover the design moments in the whole wall Ec 1 and 2 The increase in the ultimate moment resistance given by the contribution of the SFRC layer ranges from 14 5 As of the BLss 10 wall type to 21 0 As min of the BL 5 10 wall type in relation to the resistance of the SS cross section In the wall types with a 60 cm thick RC layer the above percentages are barely superior to the moment increase following the application of the second layer 1 e the increase in the cross section resistance when the SFRC layer is added is barely higher than the increase in the design moments when the second layer is considered in the structural calculation In turn the reducti
75. alls made in this thesis This section is organized in two sub sections Firstly crucial research lines are presented These lines in the author s opinion are still needed to establish this type of solution as a regular option at the moment of choosing a waterproof diaphragm wall Secondly other possible research lines which would lead to an improvement of the technique are also highlighted Bi layer diaphragm walls Experimental and numerical analysis 100 Chapter 6 6 4 1 Crucial research lines First of all more experimental results would be necessary to have more and better data to contrast models Both structural and sectional measures would be needed to contrast the FEM and AES models respectively Particularly it would be quite useful to test the bi layer diaphragm wall up to failure in order to analyze the possible failure modes and the limit states that the compound cross section can reach Secondly the differential time dependent strains creep and shrinkage between both layers was left out of the model developed in this thesis chapter 3 Although reasons were established to suppose that these effects would have a limited influence in the walls behaviour this is a strong assumption that should be checked Excessive shrinkage may lead to early debonding risk or to crack development that would be harmful for the waterproofness In this sense it would be necessary to evaluate the theoretical bond strength required for th
76. alysis 6 Chapter 1 1 3 1 Diaphragm walls A definition of diaphragm wall has already been given in previous sections together with a small description of the technique There are several books addressing this structural element e g Schneebeli 1981 Xanthakos 1979 and also a European code related to it EN 1538 2010 Diaphragm walls are classified as flexible in the group of the excavation retaining structures as its deformation influences in the general behaviour of the element and therefore in the way the element is designed In this group it can also be found the following types of walls sheet pile wall soldier pile wall also called king post wall contiguous bored pile wall and secant bored pile wall The walls can be further classified in accordance with different criteria like its structural configuration under construction e g cantilever anchored braced top down construction island excavation or zoned excavation Ou 2006 In the first of them cantilever the structure stability depends entirely on the passive earth pressures while in the others the stability is also given by one or several propping lines Furthermore the propped walls are usually sub classified in walls with one prop level or multiple prop levels The prop support system may be temporal e g bracing temporal anchors ring wales or permanent e g beams or slabs permanent anchors and its use depends on the construction sequence adopted A c
77. and repair materials Cement and Concrete Research 35 4 748 757 doi 10 1016 j cemconres 2004 05 027 Moormann C 2004 Analysis of wall and ground movements due to deep excavations in soft soil based on a new worldwide database Soils and foundations 44 1 87 98 Neville A M amp Brooks J J 2010 CONCRETE TECHNOLOGY p 442 Prentice Hall Newman J amp Choo B S 2003 Advanced Concrete Technology Processes p 699 Elsevier Ltd NLT 382 08 2008 Evaluaci n de la adherencia entre capas de firme mediante ensayo de corte CEDEX Madrid Ou C Y 2006 Deep excavation theory and practice p 532 London Taylor amp Francis Group Ou C Y Chiou D C Wu T S 1996 Three dimensional finite element analysis of deep excavations Journal of Geotechnical Engineering 122 5 337 345 Luis Segura Castillo References 107 Ou C Y amp Hsieh P G 2011 A simplified method for predicting ground settlement profiles induced by excavation in soft clay Computers and Geotechnics 38 8 987 997 doi 10 1016 j compgeo 2011 06 008 Ou C Y amp Lai C 1994 Finite element analysis of deep excavation in layered sandy and clayey soil deposits Canadian geotechnical journal 31 204 214 Retrieved from http www nrcresearchpress com doi abs 10 1139 t94 026 Ou C Y amp Lee K H 1987 Watertightness of the diaphram wall at vertical joints In Proceedings of the Eighth Asian Regio
78. ant conferences were selected for publishing papers and to disseminate the results The objectives and methods are summarized in Table 2 It can be seen that the detailed methodology is related to the objectives established Luis Segura Castillo Introduction 15 1 5 THESIS STRUCTURE In order to maximize the dissemination of results a thesis by publication was chosen Table 3 presents the list of papers that conforms the main body of thesis The following information can be seen in the table number of paper Journal indicating Impact Factor I F 2012 according to Thomson Reuters Journal Citation Reports and quartile in its category included in the footnote complete title of paper and authors Table 2 Summary of objectives and methods Objective Corroborate the viability of the proposed solution O1 Methodology Full scale experimental campaign M1 Assess the bond strength reached between the concrete layers O2 Adapt a direct shear test feasible to be used in extracted cores M2 and use it to evaluate the evolution at early ages of the bond strength M3 Assess the structural behaviour of the bi layer diaphragm walls O3 Develop M4 and validate M5 a FEM structural model Develop an overall flexural design model structural and sectional level 04 Adopt a sectional model M6 and integration with the structural model M7 into an overall design method Quantify the efficiency of
79. apter 3 0 2 and a specific weight of 24 RN m were considered for the concrete of both layers Its modulus of elasticity according to EC 2 EN 2004b was Eons 33000 MPa EA fixed end anchor I element E 10 A A i EN plate element PER K P wiinterface 4 areal S a EROS BRE SORT SER RRI MERS A CI EI RES RR RD gt SATA i FIA 4 PS DS INES ae DIT VK DAV VSO SIO AS TODS LOA DADOS PLOT CROAT OOS AA APA ll 20m A V Fig 22 Finite element model mesh and main elements The flexural ED and normal EA stiffness values calculated for the SS and the CS cross sections are shown in Table 13 All stiffnesses have been reduced by 20 from the nominal value uncracked cross section to consider the existence of cracks in the wall Khoiri amp Ou 2013 In the FEM model the self weight and stiffnesses were updated from the SS to the CS values for the corresponding beam lengths that had been sprayed after each of the spraying stages As it is very important that the ratio of El EA is not changed to avoid numerical inconsistencies PLAXIS 2D 2010a and our main interest centres on the bending moments the EA values of the bi layer were calculated to maintain a constant EI EA ratio It can be seen that the flexural stiffness of the CS for both bi layer alternatives increased by about 60 in comparison with that of the SS Table 13 Flexural and normal stiffness of the different walls
80. arate phases The first a diaphragm wall is built in the conventional manner Once this wall attains the necessary strength and after excavating the soil within the perimeter sprayed concrete with steel fibres forms the second layer The solution is schematically represented in Fig 40 Area with sprayed gt f E E a Bond plane concrete Ba LB ite I 3 Ed Vertical reinforcement Unsprayed o l A i i intrados area P h Vertical reinforcement j 0 14 extrados Y YN 1St phase reinforced Ground concrete 2d phase sprayed concrete with fibers Fig 40 Sketch of bi layer wall general and sectional view The bond between both concretes plays an important role in the performance of the structure If bond strength is sufficiently high the structure behaves monolithically effectively mobilizing all the strength of its different elements This is very necessary for the repair and reinforcement of concrete structures It is common practice first of all to increase the roughness of the base layer by applying a bonding agent and or steel connectors in some cases followed by the reinforcement layer Julio Branco Silva amp Lourenco 2006 Usual examples of this application include bridges built in several stages techniques for pavement Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 79 reinforcement Delatte Jr Wade amp Fowler 2000 and more complex techniques such as NSM
81. arge increase in the values of bending moments after the SFRC layer has been sprayed The values for M gt are much higher than the values of the MP for example in the maximum MS 519 1 KN mand MP 339 1 kN m Bi layer diaphragm walls Experimental and numerical analysis 74 Chapter 4 This situation is not desirable as on the one hand the greater the increase in the forces after the spraying the larger the shear forces in the bond plane On the other hand higher bending moments in the compound cross sections mean higher tensile stresses in the SFRC in the service state which increases the risk of crack formation and hence water filtration Finally as was seen in sections Error Reference source not found and Error Reference source not found the earlier the SFRC is sprayed the larger the increase in the bending moments The Top Down cases see Fig 39d show two areas where reinforcements for the positive moments are needed in the embedded area and at the depth where the intermediate slab is connected 6 0 m This moment is larger in the bi layer alternatives as the stiffness of the wall is higher after spraying the SFRC layer In turn the negative moments in the intermediate area of the wall depths between 6 0 m and 15 5 m are reduced This change in the bending moments is also translated into a reduction in the extra reinforcement as the negative reinforcement has to be placed across a larger section of the wall than th
82. as especiales A pesar de la distancia la vida el tiempo y el espacio tienen un lugar importante en m en alg n lugar de mi coraz n y me ayudaron a avanzar en este proyecto sabiendo que m s cerca o m s lejos siempre estaban ah Isaac desde el comienzo hasta el fin miro con ternura la manga de mi camiseta Flo una guerrera de la vida armada con su alegr a Gracias por compartirla conmigo Silvana gracias por todos los almuerzos y por guiarme en los senderos de esta existencia espiritual Vic gran compa era de la vida Fabi n creo que hablamos 3 veces en estos 4 a os pero cada una de esas veces me sent en el parque de los ni os con un rosado glamour te extra o Veronika I enjoyed and grew with you thank for all what we have lived together Leo el mejor compa ero de viaje Giulia ambarab cicci cocc l Abus gracias por estar siempre esper ndome con un caf con leche y galletitas Poly te quiero Angel Navegar e preciso viver nao e preciso Nuestros botes no van juntos pero se cruzan soplados por los mismos vientos Caro siempre ah al firme Beleza Marga que me ha sabido malcriar en cada visita aguantar todos mis llamadas pre viajes y hacerme sentir que siempre estaba cerca Lulu alma hermana gracias por estar cerca e iluminar mi vida con tus reflexiones y sobre todo con tu sonrisa alimento de mi alma Esta tesis est dedicada a la memoria de tres grandes que forjaron gran parte de lo que soy Mi viejo se
83. ater bending moments 2 2 2 Construction of experimental bi layer walls Details of the bottom up construction sequence of the experimental bi layer walls are summarized in Table 5 The following information is given for each stage a brief description number of days from panel casting to completion of the stage a reference name used to identify the inclinometer reading and the structural scheme of the model A schematic diagram of the different construction sequence can be seen in Fig 7 30 30 Details of the materials used and of the construction sequence are given below 6 1 80 3 80 Bal Tre 2 25 Bid ddd de a b c d f g h i j k Fig 7 Wall construction sequence A conventional reinforced concrete diaphragm wall constituted the first layer of the bi layer walls with a theoretical compressive strength at 28 days of f 30 MPa UNE EN 12390 3 2003 The excavation process began immediately after the cap beam had been placed in position over each complete line of panels The main characteristics of the anchorages are given in Table 6 The rods were positioned when the excavation reached the required depth Panels with anchorages alternated alongside panels with no anchorages Struts instead of anchorages were fixed to the corner panels A single line of anchorages was used around the entire perimeter except in the experimental panel of Wall W35 where two anchorages were used Following completion
84. ation Paper 1 M1 M4 M5 M8 Paper 2 M4 M6 M7 M8 Paper 3 M9 Paper 4 amp C Paper 1 M1 M2 M3 1 5 1 Chapters outlook A brief outlook of every chapter is outlined below highlighting the connection between the different papers and the methodology M previously described In each paper the methodology used is further detailed Chapter 2 Paper 1 Experimental and numerical structural analysis The structural part of the experimental campaign M1 is reported in this chapter together with the structural model M4 and its contrast and adjust M5 A first evaluation of the advantage of the walls is performed Bi layer diaphragm walls Experimental and numerical analysis 18 Chapter 1 then with the developed model M8 In this case the profitability is evaluated just until the structural level comparing the bending moments developed both in mono layer and bi layer walls Chapter 3 Paper 2 Structural and sectional analysis In this chapter the structural model developed in the previous chapter M4 is extended changing the soil model to be able to use it in a larger excavation A sectional model is adopted M6 and integrated with the structural model M7 to obtain the overall design method With the overall design method a study based on the comparison of various hypothetical cases of bi layer diaphragm is performed to evaluate again the profitability of the new wall type M8 This time the comparison is based
85. ation takes place a stabilizing fluid bentonite slurry is introduced When the excavation is complete the reinforcement cage is introduced into the trench and the concrete is cast by tremie pipes displacing and allowing recovery and recycling of the stabilizing fluid from the bottom up Although their basic principle reminds the same the technique has been developed improving the methods and equipment e g different types of dragging tools are now available drilling bit hydrophraise or hydrofrase trenchcutter hydraulic grab mechanical grab clamshell improved stability of the cutting face of the excavation reduced leakages at panel vertical joints allowance of force transmission between panels improved bracing and anchoring systems turning it into a competitive solution for deep excavation works Moreover the functionality of diaphragm walls has also widened being nowadays used as retaining walls load bearing walls cut off walls or a combination of the aforementioned Despite the technical advances the technique still presents some drawbacks One widespread problem associated with this construction technique is leakage whenever the walls are erected in water bearing ground Puller 1994 being their waterproof capacity a source of debate since the first walls of this type were built Although leakage in the walls can occur for several reasons detailed in Puller 1994 the main one is that generally the joints between panels d
86. ations there are some relatively new ones where the FRC can be partially or totally used in substitution of the conventional reinforcement It is also worth mentioning that several studies now also included in the codes demonstrated the possibility of partial or total substitution of the shear reinforcement in beams or the transversal reinforcement in thin web elements Martinola Meda Plizzari amp Rinaldi 2010 1 3 4 Sprayed concrete Sprayed concrete is a special concrete that can be defined as Mortar or concrete pneumatically projected at high velocity onto a surface ACI Committee 116 2000 Although the technique was originally patented as Bi layer diaphragm walls Experimental and numerical analysis 12 Chapter 1 Gunite it is usually called Sprayed concrete in the European influenced countries and Shotcrete in the area of influence of USA Simon Austin amp Robins 1995 Newman amp Choo 2003 can be named as book addressing this technique It is also worth mentioning as reference the following state of the art reports and recommendations AFTES 1996 Franz n 1992 ITA 1993 A brief summary of some of the salient aspects on the subject are presented hereafter The system involves spraying projecting the mix which usually has small sized aggregates at a hard surface Impelled by compressed air the material is rapidly placed and compacted even on vertical surfaces and within certain limits
87. ayer wall general and sectional view c cccscceesseceeececeeeeesaeceeaceceeeeeesaeeeeaeeeaeeeeeees 78 41 Details of diaphragm walls a site plan b side Vie W oooonocccnncccnonccononanonncononccnnnnnnnnacnnn cc cnn ccnnncnos 80 42 Main steps in the production of the specimens a milling the wall built in the first phase b finished surface c water jet washing d placing the bonding agent e area of water leakage f spraying the second layer Of concrete c oooonoccnnccnnocononcnonononcnnncnnncnnn nono nono nonn crono cono cn acc nn cnn nannnrnnn nan c conan 82 43 LCB shear test a device sketch b test configuration oooonoccnnncccnoncconnncnonnccnnnnononononnnocnnn conan co nnncnos 84 44 Core positions tor shear test yr mesier piei iia Ge 84 45 Typical shear test strength displacement graph oooooonnncccnonoconocanoncconnnncnnnocononocnnnnonnncconnncnnn conan ccnnnnns 86 46 Failure mechanisms in the test 0 iesiesontiecionctsmmcinsositzm mentisstenen lmt stemstttteed EA aea arao eas 86 47 Atypical failure a stone chip protruding on one side b On two planes eooeeoaemoaueea mena nens 87 48 Shear stress vs age of second phase concrete surface preparation with milling MP and saturated SUCEDA Ee EN sie tia 87 49 Shear stress vs age of second phase concrete surface preparation with epoxy EP 89 50 Direction of stress with regard to milling
88. aying the second layer and reproduced both the qualitative and the quantitative displacements of the instrumented walls with a high degree of accuracy 3 The model compared a theoretical bi layer wall 35 10 cm thick with two referenced single layer walls 35 cm and 45 cm in order to identify possible structural improvements The structural behaviour of both the bi layer wall and the single layer wall with the same first layer thickness was Bi layer diaphragm walls Experimental and numerical analysis 40 Chapter 2 similar in relation to their bending moments and deformation The second layer however allowed a reduction in the reinforcement required in the first layer In the theoretical case a 17 reduction in bending moments was achieved for the first layer which would result in a partial reduction in reinforcement 2 8 ACKNOWLEDGEMENTS Funding was made available from the Spanish Ministry of Education and Science through Research Project BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras and through UPC project CTT 8062 Luis Segura is grateful for the Fellowship awarded by the FPU Spanish Research Program AP2010 3789 The authors wish to acknowledge the valuable help given by Dr Chris Goodier from Loughborough University Thanks are also due to James A K Hedger for assistance with the review of earlier versions of the manuscript Luis Segura Castillo All in all it s just another brick in t
89. bi layer area work is a measure of the potential contribution of the second layer which is included in Table 17 as EA M T The area inside the M T envelopes EA M hi is also included in Table 17 It can be seen that EA M T is smaller in the bi layer than in the mono layer alternatives for all groups which is consistent with the previous results Segura Castillo Aguado de la Fuente et al 2013 Segura Castillo Aguado amp Josa 2013 The reduction ranges from 4 to 21 according to each particular case Envelope area is not directly related to the variation in steel reinforcement that can be obtained even though it is a useful indicator Different variations in steel reinforcements AVV A ex were obtained for the different cases in each group ranging from a slight increase 0 1 kg in case BL 55 10 BUs 4u 4S A to a Luis Segura Castillo Parametric study of construction processes 65 s significant drop 21 7 kg in case BL 60 10 TD 2u 2S B although the EA M d variation is similar within each group Two considerations should be noted to explain the aforementioned situation Firstly the steel reduction is limited by the contribution of the SFRC layer As can be seen in Table 17 the My increase in the cross section when the SFRC layer is added from the M S to the M 8 remains relatively constant between 43 6 kN m and 39 7 KN m in all cases and represents a percentage increase of between 8 and 15 of the
90. bond level is less connected with the other levels this can also be seen in Fig 5 To justify this it is worth mentioning two important aspects that were in the original thesis plan but could not be Luis Segura Castillo Introduction 17 included in the thesis Firstly the experimental analysis at sectional level and secondly the connection between the measured bond and a theoretical evaluation of the bond strength necessary for the correct behaviour of the walls The first aspect was not included due to different errors during the experimental campaign and the second is currently under development but could not be included simply because of lack of time before submitting the thesis Both aspects are included in the suggestions for future research at the end of the thesis Chapter 1 Introduction Chapter 2 Paper 1 Chapter 3 Paper 2 Chapter 4 Paper 3 Chapter 5 Paper 4 Experimental and Structural and Parametric study of Evolution of concrete numerical structural sectional analysis construction to concrete bond analysis processes strength at early ages Appendix 1 C Paper 1 Chapter 6 Early ages concrete to Conclusions and concrete bond strength assessed through shear and pull off tests recommendations Fig 5 Outline of the thesis Table 4 Connection between main areas papers and methodology used Paper Structural level Sectional level General design Bond Strength analysis analysis and optimiz
91. case of Wall T BLW45 are plotted in Fig 16 The maximum deformations at the top and at the centre of the wall are observed in the last constructive stage elim150 This stage will be used to perform the comparison as in the other two theoretical cases The maximum displacements in the three theoretical cases are plotted in Fig 17 Similar displacements were calculated for the two cases with the same thickness of cast layer T W35 and T BLW45 while a smaller deep inward displacement and a larger surface displacement was calculated for the thicker wall T W45 The influence of the second layer on any final displacements was not very large and may in future be calculated in accordance with the thickness of the first layer Bi layer diaphragm walls Experimental and numerical analysis 36 Chapter 2 A Eb cast exc175 anc150 exc525 exc630 elimi5C 0 0 gt 1 0 2 0 4 3 0 4 RO 4 0 5 0 6 0 Depth m 7 0 8 0 9 0 10 0 M1 1 5 0 4 0 3 0 20 10 0 0 1 0 Displacements mm Fig 16 Horizontal displacements obtained by the PLAXIS model for the T BLW45 design 00 10 4 2 0 3 0 40 5 0 6 0 Depth m elim150 70 T W35 8 0 T BLW45 90 om T W45 10 0 5 0 4 0 3 0 2 0 1 0 0 0 Displacements mm Fig 17 Maximum displacements for the three comparative cases 2 6 2 2 Bending moments Calculation of the reinforc
92. ccording to the guidelines outlined in the previous section The main drawback of the bi layer solution compared to a single layer one may be the increase in cost implied by the placement of the second layer which not only includes the material cost but also other factors such as placement logistics and design The additional drawback of an increased final thickness of the wall in the bi layer solution i e less available space inside should also be considered Nevertheless the final thickness of the single layer wall will ultimately be as thick as or thicker than the bi layer wall following the selection and the application of one of the common methods to improve watertightness In contrast the following functional advantages may be outlined for a bi layer solution 1 a wall that has improved watertightness built into its design avoiding future uncertainty over repair work due to leakage 2 even surface finish provided by the second layer 3 economical use of space especially compared with the construction of inner wall solutions and 4 improved efficiency in the use of materials through the multiple functionalities of the second layer The solution also appears particularly suitable if used on large construction sites where the switch between excavation temporary support and spraying tasks is not a significant problem 2 7 CONCLUSIONS A new structural type of slurry wall referred to as a bi layer diaphragm wall and its associa
93. ces between the mono layer and bi layer alternatives for the same structural configuration the maximum differences being in an order of magnitude of 0 1 mm Nonetheless the displacements are barely smaller in the bi layer cases due to the increased stiffness contributed by the second layer 4 3 4 Influence of the construction sequence The selection of the construction sequence in a deep excavation project depends on many factors for example adjacent excavations and constructions and their foundations construction timetable equipment and budget and the geometry and dimensions of the building Ou 2006 These factors influence the response of the bi layer technique and are analysed below Fig 37 shows the bending moment envelopes for the cases with construction sequence BUa BL 60 10 BUa 2u 2S A BUs BL 60 10 BUs 2u 2S A and TD BL 60 10 TD 2u 2S A It also includes the bending moments of six representative intermediate stages in grey and black lines which are the stages that best define these envelopes The structural configuration in both its temporary and its final stages largely determines the magnitude of the bending moment values that develop in the walls The cases that are shown here although built with different construction sequences show comparable results as all three have 2 supports during the Luis Segura Castillo Parametric study of construction processes 71 construction 2 struts and
94. ces might be due to lower precision of the experimental data at the selected reference stage evident from the trend of the offset that lay slightly outside the general trend of the data set at that stage equivalent to those included in section 4 2 for Wall W35 In confirmation of this hypothesis Fig 13 also shows the relative incremental displacement of the Control 2 stage using the Exc630 stage as a reference stage The adjustment is evidently better although the displacement is smaller Notwithstanding the uncertainties in the experimental data and the simplifications in the model good adjustment may generally be seen between both in the two walls 2 5 2 Wall behaviour Horizontal displacements and bending moments of the adjusted PLAXIS model for Wall W35 are shown in Fig 14 The expected qualitative behaviour can be observed in general throughout the different stages One result that may attract attention is the small deflection after the first excavation when the walls behave in cantilever mode which could be explained by a number of reasons A preliminary excavation of 1 0 m in depth was completed in the area prior to the construction of the walls which was also included in the model Moreover the high superficial loads 50 EN m on the other side of the street introduced to account for the nearby buildings increase the pressures and hence the displacements in the lower parts of the walls Finally the finite element mod
95. cids Wax emulsions Finely divided solids Inert Fullers earth pore filling materials Talc Bentonite Other siliceous powders Chemically reactive or Silicates SCM Finely ground blast furnace slag Pozzolans Crystalline materials Proprietary products Conventional Water reducing admixtures and Air entraining miscellaneous Accelerator Methyl siliconates Polymer The effects of the admixtures on the permeability of concrete can be evaluated both by direct and indirect measuring conductivity of chloride penetration methods A review of these methods even under loaded and cracked specimens was performed by Hoseini Bindiganavile amp Banthia 2009 As the measured permeability is strongly dependant on the test method used authors seem to agree that there is a need to standardize the test procedures in order to be able to systematize the comparisons Hoseini et al 2009 Ramachandran 1995 Finally it can be said that the required waterproofness of the walls considering the service conditions expected can be reached if an appropriate PRA is selected and if cracking of the second layer is controlled 1 3 3 Fibre reinforced concrete Fibre reinforced concrete FRC is defined by ACT Committee 116 2000 as concrete containing dispersed randomly oriented fibers Its modern development started around the 1960s after the works of Romualdi Batson and Mandel Zollo 1997 gradually increasing afterwards its
96. completed As it is the last section and its height is not over 2 m this section may be completed with in situ concrete instead of sprayed concrete There are actually three spraying stages in the 2S B and 2S C alternatives although in view of the short length of the third section they are left with the 2S cases 7 50 10 50 1S A 2S A 2S B Fig 30 Spraying discretization considered 2S C 4S A 4 3 STRUCTURAL AND SECTIONAL RESULTS Table 17 summarizes the general results for all of the cases The horizontal double line separates the cases with different final structural configurations 2u and 4u The dark lines separate the different construction Luis Segura Castillo Parametric study of construction processes 63 sequences for the 2u configurations In all cases the light line separates groups of cases vvith different first layer thickness 55 cm and 60 cm The first case in every group corresponds to a mono layer alternative with the remainder referring to different spraying alternatives for the bi layer walls Table 17 General results for all cases ats BAM BEAMS M M Wasd AW Asen Guy ASH EN mm kKNmm Nm KN m kg kg mm mm ML 60 0 BUs 4u 0S M 4518 300 2 49 2 9 73 BL 60 10 BUs 4u 1S A 4323 213 298 6 336 3 48 7 0 5 9 74 0 00 BL 60 10 BUs 4u 2S A 4180 355 298 6 336 3 46 4 2 8 9 74 0 01 BL 60 10 BUs 4u 4S A 4017 568 307 6 345 4 49 0 0 3 9 71
97. connnnnncnnonnnonnoconononccnn cana nrnn cra 119 62 Desviaci n est ndar obtenida para las distintas edades ooonncnnncnnncnnnconocnnocnonnconncconoconcconccanncnccnnnos 119 63 Resistencia a tracci n vs edad del hormig n de 2 fase para preparaci n por Fresado 120 64 Resistencia a tracci n vs edad del hormig n de 2 fase para preparaci n con adhesivo Epoxi 121 65 Esquema de irregularidades en la interfase entre capas y superficie de pantallas 122 66 Foto del acabado final del proyectado en los muros Bi Capa coconnccinccnnccnonnnooncnoncnnonnncnnncnnncnnnannnos 122 67 Ejemplo de desalineaciones sufridas durante la realizaci n del ensayo pull 0ff e 123 68 Resistencia a tracci n vs Resistencia a corte para preparaci n por Fresado oooooonoccnoccconccnoncnanannos 123 69 Resistencia a tracci n vs Resistencia a corte para preparaci n con adhesivo Epoxi e oooee 124 Luis Segura Castillo List of tables xiii LIST OF TABLES Table 1 Classification and examples of permeability reducing admixtures eee esseesseeseceseeeeeeeeeeeeees 10 Table 2 Summary of objectives and methods ee eeeeeeseesesseceseceseceseeseeeeeaeeeaeeesecaecsaecsseeeseeeeeesseeeeaeeaaees 15 Table 3 List of papers and conference papers related to the thesis 0 0 0 ee eeeeeseeeseesseceseceseceeeeseeeseeeeneeaaes 16 Table 4 Connection betw
98. d although the criteria used in the analysis are not completely realistic for example in some cases reinforcement is placed just to cover a small increase in the bending moments they allow quantification and comparison of the quantity of reinforcement steel required in the different solutions 4 2 2 Parameters under study Table 16 presents the parameters and a brief description of the alternatives that are studied As the combination of all alternatives would lead to a total of 120 cases a selection of combinations 30 cases is presented in order to analyse the influence of type of wall mono layer or bi layer construction sequence number of underground levels in the final configuration number of spraying stages and depth of sprayed concrete layer Given the large number of cases the following labels are proposed for ease of identification NL VV1 VV2 CS NU NS DS where NL is the number of layers of the wall conventional diaphragm walls referred to as mono layer walls ML for the sake of clarity and bi layer walls BL W1 is the thickness in cm of the 1st layer the conventional RC wall 55 and 60 W2 is the thickness in cm of the 2nd layer the SFRC sprayed layer O ML wall and 10 CS stands for the construction sequence where three alternatives are considered two Bottom Up BU sequences one with struts as temporary supports BUs and one with temporary anchorages BUa and the Top Down sequence TD
99. d both with an equivalent support length of 10 m with stiffnesses in the range of the Kung G T C Kung 2009 parametric analysis The Analysis of Evolutionary Sections AES model was used to perform the numerical simulation of the mechanical behaviour of the composite cross sections of the wall de la Fuente Aguado de Cea et al 2012 de la Fuente et al 2008 It simulates the non linear response of cross sections built with different materials concrete and steel and most especially the structural contribution of the SFRC under tension The characteristics of the aforementioned structural and sectional models are fully described in Segura Castillo Aguado de la Fuente et al 2013 likewise the properties of the materials were also taken from the aforementioned paper The design of the reinforcement followed the same criteria in all cases a a symmetric reinforcement at each face of the wall with the minimum reinforcement area Asmin CPH 2008 and b extra Bi layer diaphragm walls Experimental and numerical analysis 60 Chapter 4 reinforcements Asex in each point where the design bending moment M4 exceeded the ultimate bending moments M given by the reinforcement of a Only tensioned bars were taken into account in the calculation For the sake of simplicity the reinforcement was defined indicating only the necessary steel area without defining the type diameter and number of bars On the other han
100. d second layer of 10 cm the increase in the cross section ultimate bending resistance when it is strengthened by the SFRC layer i e when it changes from the simple cross section SS to the compound cross section CS is between 8 and 15 It follows that the main flexural resistance is provided by the first layer the RC diaphragm wall giving the SFRC layer a secondary flexural resistance This allows in the first place a reduction in the steel reinforcement of the first layer up to 7 0 of the total flexural reinforcement for the cases of chapter 4 Furthermore in some extend it also collaborates with a displacements reduction reducing up to 7 3 of the maximum displacements in a 20 m high wall constructed with the Top Down construction method Case BL 55 10 TD 2u 2S A of chapter 4 It was also found that the spraying sequence is a crucial parameter to be able to take advantage of the SFRC collaboration Specific indications are described in the following section The bonding capacity between layers which is crucial for the development of the technique was also analysed A good bond level was obtained in a laboratory shear test over the cores extracted from the full scale experimental walls The average shear strength value measured for each age although depending on the surface preparation was always above 1 0 MPa The measures were obtained at 2 6 and 35 days after the spraying of the SFRC layer for every preparation This is a tim
101. d from http yo 1 ct ntust edu tw ge files articlefiles v5i120100525218817485 pdf Long M 2001 Database for retaining wall and ground movements due to deep excavations Journal of Geotechnical and Geoenvironmental Engineering 127 3 203 224 Malmgren L 2007 Strength ductility and stiffness of fibre reinforced shotcrete Magazine of Concrete Research 59 4 287 296 doi 10 1680 macr 2007 59 4 287 Martinola G Meda A Plizzari G a amp Rinaldi Z 2010 Strengthening and repair of RC beams with fiber reinforced concrete Cement and Concrete Composites 32 9 731 739 doi 10 1016 j cemconcomp 2010 07 001 Meyerhof G G 1994 Evolution of safety factors and geotechnical limit state design Texas USA Texas A amp M University Retrieved from https ceprofs civil tamu edu briaud buchanan web Lectures Second Buchanan Lecture pdf Mir Recasens R Mart nez A amp P rez Jim nez F 2005 Assessing heat adhesive emulsions for tack coats Proceedings of the Institution of Civil Engineers Transport 158 1 45 51 doi 10 1680 tran 158 1 45 57833 Mirsayah A A amp Banthia N 2002 Shear strength of steel fiber reinforced concrete ACI Materials Journal 99 5 473 479 Retrieved from http www concrete org PUBS JOURNALS OLJDetails asp Home MJ amp amp D 12326 Momayez A Ehsani M Ramezanianpour A A amp Rajaie H 2005 Comparison of methods for evaluating bond strength between concrete substrate
102. dc tal 66 Fig 33 Displacements for different spraying discretizations cee ceeccescceseeescecsseceseceseceseceseceseeeaeeeaaeenaes 67 Fig 34 Bending moment envelope for cases with different depths of sprayed concrete layer a general b A ai BA cai UN RG RR ieee at eee see Rete teas 68 Bi layer diaphragm walls Experimental and numerical analysis xii Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig List of figures 35 Bending moment envelope for two final structural set ups a 4 underground levels b 2 underground levels censats te Lo dois rios anit a Mae siii Liste 69 36 Displacements for different final structure configurations 0 0 Le eee ese ene ceteeteceeeeeeeeeeaeeeaeeeaeeenaees 70 37 Bending moment envelope for different construction sequences a Bottom Up with struts b Bottom Up with anchorages c Top DOWN 00 0 eee eseeseesseenseeseeeseceeceseneesaeesseeeaeecaecnseeeeeeneeeeneeees 71 38 Displacements for different Construction SEQUENCES ee eeeeeeseeeseeeseceseceeceeeeseeeseeeeeaeeeaecaecnaeees 72 39 Design and ultimate bending moments envelope examples a efficient design b inefficient design c large increase in bending moments after spraying d Top Down case eee 73 40 Sketch of bi l
103. del hormig n de 2 fase para preparaci n por Fresado Se puede observar la gran dispersi n en los resultados obtenidos para todas las edades y en ambos tipos de preparaciones Se adjudica principalmente a dos motivos la existencia de estas grandes dispersiones Por un Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 121 lado al gran espesor de la 2 capa de hormig n teniendo en cuenta el espesor previsto en el dise o y por otro a las desviaciones angulares registradas entre elemento de extracci n eje del testigo y direcci n perpendicular al paramento Como se ha comentado en la introducci n de este art culo ambos sucesos son comunes en este tipo de ensayo pero por los motivos que se indican a continuaci n parecen agravarse notoriamente en este caso concreto 0 8 0 7 0 6 O 0 4 0 3 O 8 0 2 1 10 100 Edad del hormig n de refuerzo d as 0 1 Resistencia a tracci n c MPa oo Fig 64 Resistencia a tracci n vs edad del hormig n de 2 fase para preparaci n con adhesivo Epoxi El espesor previsto en dise o para la segunda capa de hormig n era de 10cm Por las heterogeneidades propias del sistema de proyecci n resulta dif cil lograr un espesor homog neo dependiendo en gran medida de la habilidad del operario realizando la tarea Como se puede observar en la Table 23 ver Anexo los espesores registrados para la segunda capa van desde 8 5 cm
104. dered in the analysis The increase in the E value after adjustment of the model is in accordance with results published elsewhere Ou amp Hsieh 2011 which point to high soil stiffness at small deformations Plate structural elements were used to model the diaphragm walls A specific weight of 24 kN m was considered for the concrete Its modulus of elasticity was calculated from the compressive strength fem 28 according to standard EHE 08 CPH 2008 where Eemos 27756 MPa The elastic modulus time evolution was ignored as its difference was below 4 see section 4 1 The same elastic modulus was also considered for both concrete layers as their values are relatively similar for both concretes see section 4 1 and the change in the bi layer stiffness is largely due to the increase in cross sectional thickness rather than any change in the elastic modulus The initial thicknesses considered in the analysis for both layers were their theoretical design values i e 35 and 10 cm for the first and second layers of Wall W35 and 45 and 10 cm for Wall W45 respectively The final values obtained after the back analysis are shown in Table 8 These values are reasonable considering for the first layer that an increase in the dig hole may be produced by the digging process and for the second layer the margin of thickness observed in the extracted cores Table 7 Geotechnical parameters used in the PLAXIS model Level Type of soil Y
105. ds on the experience of the worker applying it which implies an additional factor adding more variation to the results Notwithstanding the scattered data a decrease in strength with age is observed Tu Tu amp Kruger 1996 also noted a decrease in strength after 14 days attributing it to the deterioration of epoxy caused by water that migrated from the fresh concrete and gradually accumulated at the epoxy concrete interface Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 69 2 00 1 80 HH LL 1 60 HH LL 1 40 A LL LL 1 20 HH HH 1 00 HH A LL 0 80 HH HH 0 60 HH 0 40 HHH HH 020 Hi Lael 0 00 T i 1 10 100 Age of 21d phase concrete days Shear strength t MPa Fig 49 Shear stress vs age of second phase concrete surface preparation with epoxy EP 5 3 2 5 Milling direction Fig 50 shows the graph for shear stress according to the age of second phase concrete at the time of testing grouped according to the direction in which the load was applied with regard to the milling direction This graph only considers the results of the MP case Conflicting results were obtained for different ages At 2 days the strength of the HM case was higher at 6 days the results alternated and at 35 days the strength of the VM case was higher 1 80 1 60 HH 2 1 40 A Q 1 20 A x LL e x E 1 00 o A LL 2 x
106. e but incorporates a water table at depth 5 0 m The construction sequence was also modified adding the lowering of the water level described above The second new case Loose Sand is similar to the BL 60 10 BUs 2u 2S A case but the properties of the soil were modified to reflect a looser sand the elastic parameters Esp Eseg and E were reduced by 50 and the q by 5 The corresponding mono layer alternatives for each of the previous cases were also simulated The main results are summarized in Table 18 where the same information as in the previous cases is provided In the Water Ground case a steel reduction of 8 5 kg is obtained Although the reduction is not large it is a relevant improvement compared with the base case BL 60 10 BUs 4u 2S A where the reduction achieved was of only 2 8 kg due to the contribution of the SFRC layer that withstands the moment increase caused by the water table level recuperation The reduction may even be improved if an adequate spraying sequence is selected Luis Segura Castillo Parametric study of construction processes 75 Table 18 General results for the additional cases EM BAM BEAMS M M Wasd ANA Sumo ASH kNmm kNmm kNm KN m kg kg mm mm ML Water Ground 3737 288 0 30 0 8 85 BL Water Ground 3361 403 259 0 296 3 21 5 8 5 8 81 0 04 ML Loose Sand 6168 514 9 336 3 139 1 16 70 BL Loose Sand 5413 722 477 9 517 5
107. e positive one 4 4 ADDITIONAL CONSIDERATIONS A diaphragm wall civil engineering project involves many parameters The classification by Kung G T C Kung 2009 lists some of them inherent parameters e g stratigraphy and site environment design related parameters e g properties of retaining system excavation geometry and strut prestress and construction related parameters e g construction methods over excavation and prior construction Although the analysis developed here was limited to the study of 6 parameters it provides the basis for the understanding of the behaviour of other parameters that are not included for example if a stratigraphy with an elevated ground water table is considered In this case a common construction technique consisting of lowering the water table during the excavation to avoid soil liquefaction may be used Once the excavation and the substructure have been completed the water table may be restored to the original value increasing the loads on the wall This increase can also be covered by the extra strength provided by the SFRC layer as was done with the increase caused by the internal redistribution of forces described in section 4 3 In a similar way the analysis can be extended to other type of soils To quantify this four additional cases were simulated comparing mono layer and bi layer alternatives The first new case Water Ground is similar to the BL 60 10 BUs 4u 2S A cas
108. e built in the 1950s and 1960s their potential waterproofness has been widely discussed and several techniques have emerged to prevent the emergence of leaks or repair them Puller 1994 A standard technique for repairing leaky walls is to repair the affected areas chipping away the damaged element and restoring it with a waterproof mortar Leakage usually only becomes apparent over lengthy periods and it appears at different times in different areas of the walls meaning that repair works may often extend over indefinite periods of time and require several sessions A solution that is less widely used consists of casting a second layer of waterproof mortar or concrete over the inner face of these walls Since the whole surface is covered this is an effective albeit expensive solution Wong 1997 An example of this solution was used by Li in the study of tensile creep in concrete at early ages Li et al 2008 Part of a larger research project this study aims to maximize the functional attributes of the second layer of concrete by allowing it to play a structural role in addition to its initial intended purpose waterproofing In view of the structural role of the second layer the thickness of the first layer may be reduced The dimensions of this bi diaphragm wall and its watertightness make it a feasible structural solution In this way the bi layer diaphragm walls are made of two concrete layers poured and sprayed respectively in sep
109. e characteristics of the anchorage are identical to those described for the experimental case of Wall W35 Table 10 Sequence of consiruction stages of theoretical walls Reference name Description of the stage T W35 T BLW45 T W45 Casting of panels Cast Cast Cast Excavation before anchorage activation Exc175 Exc175 Exc175 After anchorage activation Anc150 Anc150 Anc150 T BLW45 Intermediate excavation Exc525 T BLW45 Spraying of second layer Spraying End of excavation Exc630 Exc630 Exc630 Construction of base amp slabs Base amp Slab Base amp Slab Base amp Slab Anchorage elimination Elim150 Elim150 Elim150 T BLW45 specified when stage differs in other cases Central to the bi layer wall design is its capability to withstand the moments that develop during the final stage of the excavation The application of a sprayed layer during the construction stages gave the compound section greater strength to resist these forces as they developed This situation is no longer the same as in the previously described experimental cases in which the compound section was tested through the addition of an auxiliary anchorage 2 6 2 Results for theoretical cases This analysis centres on two usually crucial factors in the design of the walls on the one hand the deformations and on the other hand the bending moments that develop in the walls 2 6 2 1 Displacements The displacements that the model generated in the
110. e che cbid va ties Fa tics air da Tae dee tada da ad nde dnd ri cant 59 4 2 2 Pdrdmeters understudyiis iii RON 60 4 3 STRUCTURAL AND SECTIONAL RESULTS c cccccssececsecceceseeeeceeaeececeeneeceesaeeeseaaececsuneeceesueeeceeaaeeseeeeceeaueeeceeaaeees 62 4 3 1 Influence of the number Of spraying stages ccceccccecssscccsessscscesceceessececsesssseseeseecessssecsesseseesesseceessseceees 65 4 3 2 Influence of the depth of the sprayed concrete IAYVEL ccsssccceesscceesssseeessscesseseeceesssseesessseeessseceesssecees 67 4 3 3 Influence of the final structure configuration cccccccccceessssecescecsesssceceesssseceessecesssseceessesscesseseetsseeeees 69 4 3 4 Influence of the construction sequence u ccssscccceescceessssecessssssnseecsessseceesssscssseseessssecsessesscesseeessseeeees 70 43 5 SECON MA NN ANT 73 4 4 ADDITIONAL CONSIDERATIONS ccccssccccessececsececsesseeecseceececeencecesaeeecseaaececeaeecesaeecseaaececeaeeceeueeeceeaeeeseeaaees 74 AL CONCLUSIONS EG A a AS 75 4 6 ACKNOWLEDGEMENTS 20 A A A A Roman 75 CHAPTER 5 EVOLUTION OF CONCRETE TO CONCRETE BOND STRENGTH AT EARLY AGEG ssssceeeceeeeeees 77 5 1 a A masa caves e ENESE PAREEK eA deb tude cnevandldasegete c est a REE Ae EAE a aRar ei eases 78 5 2 EXPERIMENTAL PROGRA Mii tit ie admi ta 80 5 2 1 Preparation Of SPECIMENS iii aia 81 LE A NA AAN 83 5 32 RESULTS AND DISCUSSION sc ssdects cgeccvecheccd siccsscuscassec
111. e correct behaviour of the walls and compare it with the measured bond strength Additionally although there are additives to make the concrete waterproof a real measure of it would be needed for this application It should be checked that a sufficient level of waterproofness can be achieved by the SFRC layer with the cracking levels expected after the differential time dependent strains developed and the external forces were applied on the wall Finally as it was seen the analysis of final materials use revealed a similar material consumption for the different waterproof systems It is interesting to perform a complete sustainability analysis economic social and environmental to obtain a precise evaluation of the complete cost that allows a comparison of the different waterproofing systems 6 4 2 Other research lines It would be interesting to extend the parametric analysis to other relevant variables For example all along the thesis a thickness of 10 cm was considered for the SFRC layer It may be interesting to evaluate the viability of using other thickness or even a variable thickness for different heights of the wall Also some aspects that are currently being investigated in diaphragm walls may also be investigated in the bi layer diaphragm walls For example the 3D structural behaviour which may be influenced in the bi layer case by the horizontal connection provided by the SFRC layer Or the way of linking the walls with other
112. e lapse in the order of the needed for the wall construction Although the bonding strength depends on the test chosen to measure it the magnitude obtained with the one used in this thesis is according to preliminary estimations an order of magnitude bigger than the shear stresses between layers that may be produced by the external forces On the other hand it is necessary a deeper study regarding shrinkage and creep in the second layer which may lead to the need of improving the bonding to avoid the possibility of debonding of the second layer Besides the shear test carried out a monolithic behaviour was observed at global level in the experimental walls Luis Segura Castillo Conclusions and future perspectives 97 The material consumption concrete and reinforcement steel of two bi layer diaphragm walls was also compared with an equivalent mono layer diaphragm wall combined with an added waterproof system It was found that the material used in both solutions was similar in the best case or larger in the bi layer diaphragm walls see chapter 3 Considering in addition that the technology to build it is more expensive i e SFRC instead of RC and sprayed concrete instead of sprayed mortar it follows that the construction costs should be higher in the bi layer diaphragm wall technique However it is an interesting option under particular circumstances like space limitations or if continuous maintenance costs due to drains and pump sh
113. e o Ko Zini Rinter Vv Eo Efinat EN KN m m KN m KN m 0 Heterogeneous fill 17 75 5 00 25 00 0 577 0 00 0 67 0 4 8000 50000 A Brown silty clay 18 50 10 00 27 50 0 538 3 25 0 67 0 4 40000 70000 B Debris packageof 1875 0 01 31 50 0478 7 25 0 67 0 3 90000 70000 brown sand Brown ochre and C 18 60 13 50 28 00 0 531 16 25 0 67 0 2 90000 80000 grey marly clay The stiffness values in the model are shown in Table 8 It can be seen that the flexural stiffness of the bi layer wall is almost double that of the simple layer wall Noting the PLAXIS recommendation it is very important that the ratio of EI EA is not changed since this will introduce an out of balance force PLAXIS 2D 2010a our interest centres on the bending moments that develop and the deflections that they cause The calculated EI values for the bi layer section were used and the EA values of the bi layer were calculated to maintain a constant EI EA ratio In the PLAXIS model the wall stiffnesses were changed from the simple section to the bi layer section after the spraying stage Table 8 Thickness and flexural and normal stiffness of the walls used in the PLAXIS model tinai M El MN m EA MN 1st layer 2nd layer Simp Bi layer Simp Bi layer W35 0 450 0 115 210 7 417 1 12 490 24 721 w45 0 550 0 140 384 8 759 8 15 265 30 142 Value calculated to keep the EI EA ratio unchanged Luis Segura Castillo Experimental and numerical st
114. e on the results obtained To that end the average angle of the failure plane in the load application direction was measured om in Fig 52 For its calculation the longitudinal difference between the upper and the lower points of the core marked as t in Fig 52 was measured Then Om was calculated by means of the following trigonometric expression Om arcte t dm 10 where m is the average diameter of the contact surface in mm and t is the longitudinal difference previously described also in mm Fig 53 shows the graph for the shear strength value in relation to the failure angle of the core in the MP series In these results there is a noticeable tendency which fits in with the expected model based on the concept of the bonding envelope Simon Austin et al 1999 The strength increased slightly for decreasing values of Om Positive angles combine shear with tension decreasing the value of the strength needed to reach failure stress Negative angles combine shear with compression increasing the value of that strength In any case the influence of the age of the second phase concrete is greater than that of the angle of the failure plane Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 91 core axis im failure plane ra 1 phase concrete Fig 52 Failure plane angle In the future with more experimental data if this tendency is confirmed a function could be dete
115. ears over long time periods at different times and in different areas of a wall causing problems for both owners and contractors A less widely applied solution consists of casting a second layer of waterproof mortar or concrete over the inner face of the walls Since the whole surface is covered this is an effective albeit expensive solution Wong 1997 Finally another common practice already standardized in British construction codes BS 8102 1990 2009 is to construct an inner wall separated by a cavity Puller 1994 at the bottom of which the water is left to accumulate before it is pumped out Although dry inner walls are still constructed this solution presents some drawbacks the inner wall loses significant volume in view of the cavity and construction tolerances and it may at worst conceal dangerous leakages and even structural problems The major aim of this research project is to maximize the functional attributes of the second layer of concrete based on the second lining solution described above by allowing it to play a structural role in addition to its initial intended purpose waterproofing In accordance with the structural role of the second layer the thickness and reinforcement of the first layer may therefore be reduced The dimensions of this bi layer diaphragm wall and its improved watertightness suggest that it could be a feasible structural solution Thus the bi layer diaphragm wall represents a new type of slu
116. ectado con fibras e impermeabilizante que suma a las caracter sticas propias de los muros pantallas contenci n del terreno y capacidad portante la propiedad de ser impermeable de por s Un esquema de la soluci n se puede ver en la Fig 56 r A A IA III II IA Zona con hormig n ya L proyectado Zona aun sin proyectar v nculo entre hormigones armado vertical intrad s armado vertical trasd s 1 Fase Terreno Hormig n armado 2 Fase Hormig n con fibras proyectado Fig 56 Esquema de la soluci n por muro Bi Capa Vista general y vista de una secci n La soluci n se basa en la idea ya descrita de realizar una 2 capa en todo el interior del paramento El elemento innovador consiste en que se espera maximizar la utilizaci n de la segunda capa de hormig n d ndole un fin estructural adem s de la finalidad original impermeabilidad con la que fue pensada El aporte estructural de la segunda capa se brindar mediante la utilizaci n de hormig n fibro reforzado FRC en su construcci n De este modo se espera poder reducir el espesor de la primera capa lo suficiente para volver viable la soluci n conjunta A1 1 3 Adherencia entre hormigones El desempe o del v nculo entre ambos hormigones cumple un rol importante en el desempe o del conjunto estructural Si se logra suficiente adherencia la estructura reforzada se comporta monol ticamente siendo los materiales efectivamente movilizados
117. ed with the excavation depth therefore the deeper the excavation when Ss the second layer is sprayed the larger the Ma envelope This includes the value of the local maximum moment see arrow 1 in Fig 34b In the extreme case when the SFRC is sprayed after the excavation is completed depth 12 5 m the value of the maximum M E coincides with the value of m meaning that there is no bi layer area work at all 0 1 2 3 44 5 6 7 8 E 9 10 Z 11 O 12 350 300 250 200 150 100 A B 14 DS A 12 5 m MG s Es mss 15 4 16 DS B 11 5 m Mas mss 17 18 DS C 10 5 m Mis MS p 20 O MSSmax before spraying 400 300 200 100 0 100 200 300 O Depth of sprayed layer Bending Moments RN m Fig 34 Bending moment envelope for cases with different depths of sprayed concrete layer a general b detail Secondly there is a rise in the values of the M 2 envelope located at one extreme of the intermediate sprayed section and produced after a subsequent increase in the bending moments This rise is not desirable in the position where the maximum moment develops as the simple cross section should be designed to resist that moment regardless of the extra strength provided by the second SFRC layer So it is convenient to spray the concrete at a lower excavation depth so as to move the rise away from the position where the maximum moment develops see arrow 2 i
118. een main areas papers and methodology used cee eeeeseeeseecneceseceeeeeeeeeeeeeenee 17 Table 5 Sequential stages of wall construction 00 eee eee eseencesseceteceteceseeeseeecaeeeaeecsecaeceaeceseeeeeseeeseneeeaeenaees 25 Table 6 2 Anchorage properties sesio inire oren area eaa EEs ie aa E EuS Onara ESEE E I ETa a 25 Table 7 Geotechnical parameters used in the PLAXIS modelo eee eeceeseeseeeneecnseceseceaeceeeeeeeeaeeeaeeeaaes 28 Table 8 Thickness and flexural and normal stiffness of the walls used in the PLAXIS model 28 Table 9 Compressive strength of concrete in both phases 0 cee ceeceseesseeeneeececeeceseceaecsseeseeeeeeeeaeeeaaeenaes 29 Table 10 Sequence of construction stages of theoretical WallS onnnonncninnnncinncnnoccnoncnonoconoconcconorancrnn ccoo 35 Table 11 Main results of theoretical comparison performed with PLAXIS model eee eeeeeeeeeeeeereeeeee 38 Table 12 Construction stages SCQUeNncCe eeeeeeecssecsseceseceseceseceseceseceeeeaeeesaecsaecsaecsaeceaeceseseneeeaeeeaeeeaaeenaees 45 Table 13 Flexural and normal stiffness of the different wallS o onicnnnnnnnnnccnnocnnoncnoncnnncconocnncnanc ran crnn conan 46 Table 14 Reinforcements and My of the different wall types oonncononccincconcnnoconanononcnonocnoconoconaconc cnn cono conan 51 Table 15 Comparison of different waterproofing Systems oooocococonococonoconcconananncnn ccoo nono nc nn nono noconn
119. een that the case with 4 spraying stages BL 60 10 BUs 4u 4S A see Fig 32c follows the same pattern the greater the spraying discretization the more work done by the bi layer area and rises to 12 In contrast it can be seen that in general the maximum negative moments are somehow larger in absolute terms in cases with greater discretization of the spraying stages This behaviour is logical considering that the moment increase throughout the excavation is greater in the sections already sprayed due to the Luis Segura Castillo Parametric study of construction processes 67 increase in cross sectional stiffness Nonetheless the maximum envelopes are quite similar in the three cases The greatest difference between them is 9 8 kN m depth 9 5 m comparing the 4S case with the SIS which represents a percentage difference of 3 7 Fig 33 shows the horizontal displacements for the corresponding cases shown in Fig 32 The displacements of the corresponding mono layer case ML 60 0 BUs 4u 0S M are also included rar ae E TF O ee OnN DU PWN mu O 1 o Depth m NS 0S 77 2 NS 1S 184 NS 2S 19 NS 4S 11 10 9 8 7 6 54 3 2 1 0 Horizontal Displacements mm Fig 33 Displacements for different spraying discretizations The displacements are practically identical in all cases with maximum differences of about 0 1 mm This is because the stiffness increase given by the sec
120. el and the Mohr Coulomb elasticity model overestimated soil decompression during the excavation reducing horizontal pressures in the interior soils thereby causing large scale horizontal displacements at the bottom of the walls towards the interior The reason for placing the auxiliary anchorage in Wall W35 is now evident The change in bending moments from stage exc630 to elim350 is indicated by a hatched area in Fig 14 This increment Bi layer diaphragm walls Experimental and numerical analysis 34 Chapter 2 which was activated in a controlled manner through the anchorage release was resisted by the compound section 0 0 0 0 1 80 10 4 10 Bi layer 2 0 20 3 0 4 3 0 cast exc180 anc150 _ 40 4 0 50 5 0 A 501 6 0 exc380 anc350 exc630 7 0 7 0 4 8 0 4 8 0 9 0 9 0 10 0 10 0 j d 5 50 40 30 20 10 00 10 100 80 60 40 20 0 20 40 60 elim350 elim150 Displacements mm Bending moment kN m Fig 14 Horizontal displacements and bending moments obtained by the adjusted PLAXIS model for Wall W35 The corresponding Horizontal displacements and Bending moments of Wall W45 are shown in Fig 15 A comparison of these plots with those for Wall W35 reveals the difference in stiffness of both walls While the displacement of Wall W45 differs by almost 2 mm from the top to the centre Wall W35 has differences of more than 3 mm 3 0 4 0 E P 50 5 0
121. els where used In a first instance a Winkler model was used with two purposes firstly to design the experimental campaign and secondly to perform a preliminary parametric study in order to identify the main parameters Afterwards a FEM based model was developed to model the walls and the soil A comparison between models was performed after which it was decided to use the FEM model to perform the subsequent analysis The structural experimental results were analysed and the FEM model was contrasted and adjusted with them A sectional model capable of modelling the different materials present in the compound cross section including the SFRC was adopted The AES model was chosen because it met the mentioned requirements Both structural and sectional models were integrated to establish an overall design method The profitability of the bi layer walls was assessed using the overall design method The evaluation was performed comparing bi layer diaphragm walls with equivalent alternatives of mono layer walls Both levels of analysis structural and complete design were used in the comparisons The overall design method was then used to perform a parametric analysis The effectiveness of different construction sequence walls configuration and spraying sequence alternatives was studied in the parametric analysis It focused on the first layer steel reinforcement and the displacements reduction High level refereed journals as well as relev
122. ema habitual en la construcci n de pantallas continuas en terrenos con presencia de agua es la existencia de filtraciones El objetivo de esta tesis busca resolver este problema mediante el desarrollo de un nuevo tipo de pantalla la pantalla bi capa BL por sus siglas en ingl s El m todo para construir estos muros se basa en una soluci n existente realizar una segunda capa de hormig n impermeable sobre los muros pantalla En las pantallas BL la segunda capa se realiza con hormig n con fibras de acero SFRC proyectado sobre las pantallas convencionales llamadas ML en esta tesis e incluyendo a su vez una adici n impermeabilizante La idea central es maximizar las funciones de la segunda capa asign ndole un rol estructural adem s de la funci n impermeabilizante La metodolog a propuesta se basa en la combinaci n de trabajos experimentales y herramientas num ricas Se propone un m todo de dise o para las pantallas BL basado en modelos estructura secci n desacoplados Posteriormente se utiliza este m todo para realizar diferentes comparaciones con pantallas ML y un an lisis param trico exhaustivo de distintos procesos constructivos involucrados en la construcci n de las pantallas BL La campa a experimental realizada comprende dos niveles a nivel elemento se estudi la respuesta estructural de pantallas construidas en un edificio real ubicado en Barcelona a nivel seccional se midi el nivel de adherencia entre ambas capas de ho
123. ement is not discussed here for the sake of brevity Instead in a simplified approach the comparison will be performed directly with the bending moment diagram The bending moments in the case of Wall T BLW45 are plotted in Fig 18 From the first stage up until stage exc525 the wall only consists of the cast section After this stage the second layer is sprayed from the top of the wall to a depth of 5 0 m Therefore the contribution of the compound section can be considered in that area in the stages after exc525 A slight increment in bending moments at depths of between 3 0 m and 5 0 m can be seen in Fig 18 when moving from stage exc525 to stage exc625 Subsequently when the anchorage is released the positive bending moments around the anchorage point turn negative when it is eliminated creating a new area of bending moments from the top of the wall to a depth of 3 0 m Luis Segura Castillo Experimental and numerical structural analysis 37 After spraying the second layer the compound section is able to resist the bending moments that are generated which allows us to consider the collaboration of the second layer of sprayed steel fibre concrete in the aforementioned stages Bi layer ao Work 2 0 3 0 4 0 5 0 6 0 4 cast exc175 anc150 Depth m 5 25 7 0 8 exc525 exc630 elim150 9 0 60 20 20 40 Bending moment RN m Fig 18 Bending moments obtained by the PLAXIS model f
124. ending reinforcement The spraying sequence is a relevant parameter in the design of the bi layer walls In general terms in cases with no increase in moments following the excavation process the SFRC should be sprayed during the excavation if full advantage is to be taken of the strength increase given by the SFRC layer Otherwise spraying must be done after finishing the excavation process In each case the sooner the SFRC is sprayed the larger the bending moments that are recorded The displacements which are governed by the thickness of the first layer are practically identical for each combination of final structural configuration and construction sequence Although a displacement reduction is registered when the second layer is included it is minor compared with the total displacements The maximum reduction obtained 0 6 mm represented a percentage reduction of 7 3 ACKNOWLEDGEMENTS Funding was made available from the Spanish Ministry of Education and Science through Research Project BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras and through UPC Bi layer diaphragm walls Experimental and numerical analysis 76 Chapter 4 project CTT 8062 Luis Segura Castillo is grateful for the Fellowship awarded by the FPU Spanish Research Program AP2010 3789 Luis Segura Castillo Crazy but that s how it goes Millions of people living as foes Maybe it s not too late To learn how to love And
125. eneral references covering this subject which main aspects are included hereafter The PRAs have to be used in well proportioned concrete mixtures and a w c ratio below 0 45 is recommended for a waterproof concrete The PRAs are usually divided in two subcategories depending on whether the concrete is intended to resist non hydrostatic conditions called PRAN according to ACI or damproofing according to Ramachandran 1995 or if it will be exposed to hydrostatic conditions called PRAH or waterproofing respectively As the range of PRAs is so wide it is difficult to comprehensively classify these products Moreover these products usually improve other characteristics of the concrete e g drying shrinkage chloride ion penetration freezing and thawing resistance and autogenous sealing and at the same time there are many admixtures designed for other porpoises that are also able to reduce the permeability of the concrete ACI divides the PRAs in the following main families a Hydrophobic or water repellent chemicals b Finely divided solids and c Crystalline materials Ramachandran 1995 presents an extended classification differentiating the finely divided solids into reactive and inert and adding the conventional admixtures that are able to reduce permeability Accordingly ACI indicated that some authors included the supplementary cementing materials CSM among the finely divided soils Within the CSM the Cond
126. ensed Silica Fume CSF also known as Silica Fume or Microsilica shows particularly high performance reducing permeability and improving the durability of the concrete Chan et al 1999 The complete classification and a list of examples can be seen in Table 1 An alternative classification can be seen in Chan et al 1999 The hydrophobic materials work causing a reversed angle on the water solid interface forcing the water out of the pores This may be enough protection only if there is no hydrostatic pressure and if the concrete has no significant cracks Finely divided solids significantly reduce permeability increasing the concrete density or Bi layer diaphragm walls Experimental and numerical analysis 10 Chapter 1 by voids filling Accordingly both previous products are usually categorized as PRANs Crystalline admixtures have active ingredients that react with the free water and cement particles in the concrete forming calcium silicate hydrates needle like crystals that block capillaries and microcracks even generated over the life of concrete according to crystalline admixtures manufacturer crystalline treated concrete is able to self seal cracks up to 0 5 mm As the generated protection is able to resist hydrostatic pressure up to 120 m of head are registered it can be categorized as a PRAH Table 1 Classification and examples of permeability reducing admixtures Group Example Water repelling Soaps materials Fatty a
127. ent use of space especially below ground level Bobylev 2006 Expanding fleets of vehicles require the adaptation of their transport systems for circulation and parking Urban metro systems and road tunnels help to reduce traffic congestion and to minimize contamination The excavation works that these structures require should not adversely affect existing infrastructure and should minimize any interruption to the daily life of the city In this scenario the conventional diaphragm wall technique frequently represents a viable solution Economies in a diaphragm wall project may be achieved at the beginning of the design process when selecting the method the construction sequence and the type of wall and in the optimization of the temporal and permanent use of the retaining structure Gaba et al 2003 Accordingly material consumption the final dimensions of the wall maintenance requirements and construction complexity should all be evaluated before the adoption of any one solution Gaba et al 2003 A widespread problem associated with this construction technique is leakage whenever the walls are erected in water bearing ground As there are no existing techniques to make diaphragm walls fully watertight a variety of alternatives have been developed to cope with the leakage problem Puller 1994 A common technique is repairing locally with a waterproof mortar render over areas where leakage is detected However leakage usually only app
128. ents were taken according to standard practice Dunnicliff 1993 at different depths i from 9 5m depth to ground level separated 0 5 m each at every critical structural stage t These stages are shown in Table 5 in bold Extra measurements were also performed at an approximate depth of 5 m in the excavation as it was not uniform it had different depths in the different areas of the construction site and in the anchorage release the auxiliary anchorage in Wall W35 was released in two stages both of which were measured as the two cables that formed the anchorage were cut one by one Three measurements were taken for the main anchorage release in order to assess the effect of any possible drag on the side panels i e one measurement was taken when the anchorage of the instrumented panel was released and one for each of the side anchored panels Two extra control readings were performed at one and at three days after the main anchorage releases to monitor the effects of possible time dependent behaviour of the soil At each depth i and stage t two readings were taken one in the main direction and then another repeated at the same point with 180 degree rotation the probe The standard measurements of Check Sum ChS Incremental Displacement If and Cumulative Displacement At were then calculated Dunnicliff 1993 Luis Segura Castillo Experimental and numerical structural analysis 27 2 3 NUMERICAL MOD
129. er diaphragm walls Experimental and numerical analysis 44 Chapter 3 the latter it has to be mentioned that the suitability of this assumption has previously been studied in Segura Aguado 2012 3 3 CHARACTERISTICS OF THE WALLS 3 3 1 Geometry and construction sequence This study is based on the hypothetical case of the construction of walls designed for use in a four level basement The comparison considers one conventional diaphragm wall referred as the mono layer wall ML for the sake of clarity and two bi layer walls BL differentiated only by their thicknesses all other properties remaining constant e MLgo Conventional RC diaphragm wall of 60 cm thickness e BL 60419 Bi layer RC wall with a thickness of 60 cm onto which a 10 cm thick SFRC layer is sprayed e BLss 10 Bi layer RC wall with a thickness of 55 cm onto which a 10 cm thick SFRC layer is sprayed The general characteristics of the walls used in this study are similar to those used by Carrubba amp Colonna 2000 in order to contrast our results with others from the technical literature Apart from some minor differences a major difference is the increase in the penetration depth of the walls This change is because the one in the reference is below the usual range for this depth of excavation Long 2001 The selected diaphragm wall was 20 0 m high and required an excavation depth of 12 5 m with a 7 5 m embedded footing as illustrated in Fig 21a During
130. eramos que la correlaci n analizada en este apartado debe ser tomada como un resultado meramente orientativo siendo necesario repetir la experiencia tomando precauciones especiales para el control de la perpendicularidad y el espesor de capa para poder extraer conclusiones generales sobre la relaci n entre ambos ensayos A1 6 AGRADECIMENTOS Los autores agradecen a PERMASTOP TECHNOLOGIES por la financiaci n brindada CTT 8062 los recursos materiales y la asistencia de su equipo especialmente Ra l Suarez y Tom s Dur n para el desarrollo de este proyecto Tambi n a la financiaci n brindada por el Ministerio de Educaci n y Ciencia a trav s del proyecto BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras L S agradece al Programa de FPU del Ministerio de Educaci n por la financiaci n para la realizaci n del doctorado AP2010 3789 y al Instituto de Estructuras y Transporte de la Universidad de la Rep blica Uruguay por el constante apoyo recibido Bi layer diaphragm walls Experimental and numerical analysis 126 A1 7 Table 23 Resultados ensayo pull off ANEXO Resultados experimentales del ensayo pull off Appendix 1 E E an RE S Sa E 3 a ze 3 a mye da gt gt Z ge E lt 8 d as cm Kg cm2 Kg cm2 GA 35 t1 35 PE Tl 2 Rech R S 191 0 20 9 55 GB 35 t1 35 PE Tl 2 Rech RS 98 0 20 4 9 GE 35 t1 35 PE T1 2 Acept R L 15 0 210 9 41 51 5 08 GD 35 t1 3
131. es are practically identical 4 3 3 Influence of the final structure configuration The bi layer diaphragm walls have a different structural response depending on the final structural configuration This aspect is studied further by developing the analysis in the previous sections all with NU 4u for its application to a case with a different final structural configuration NU 2u The bending moment envelopes for a case with 4 underground levels BL 60 10 BUs 4u 2S A and a case with 2 underground levels BL 60 10 BUs 2u 2S A are shown in Fig 35a and Fig 35b respectively 0 0 1 1 2 2 3 3 4 4 4 5 5 6 6 7 7 84 8 E 9 9 T 0 10 5 u 411 A 12 12 13 13 14 14 15 15 16 4 NU 4u 16 NU 2u 17 Final Exc 17 Final Exc Supp Out Supp Out 18 mss 1814 mss o 194 MP al 194 MS b 20 4 20 400 300 200 100 0 100 200 300 400 300 200 100 0 100 200 300 Bending Moments RN m Fig 35 Bending moment envelope for two final structural set ups a 4 underground levels b 2 underground levels As the excavation stages are identical for both cases the envelopes differ at the stage when the temporal supports are removed and the walls are supported by the finished slabs As seen in section 4 3 1 in the 4u case see Fig 35a the internal redistribution of forces is small and the values of the maximum bending mo
132. esccceseddiseciueccacdeceseesssdoaccdessdiessdiestccecedsest desaccdossdesecsiensecesacd VII LIST OF FIGURES fossccditscicccecseccheeccececscseccsdedeces occcesecsescecsseacccecaasscdceccccesssededcccecdces cedssecdentuscccessaaecess cccdsescesecvecedcecedsces XI LIST OF TABLES aaa XIII CHAPTER 1 INTRODUCTION wssiccssiccceccccscicdsacesecustnesticccctecdedestidesceccaddacecdaceucceceshsevieeeotcdecvsucedacdeced socadesddaceccesecess 1 1 1 MOTIVATION oca eo dte a de dien ae desea eee 1 1 2 THESIS OBJECTIVES vrniti citi ram iria Biada edacdechanbadeveuesthihbedecaevesdodassbenedessedachbeseceevanticdanse 4 13 THESISBACRGROUND AA EAN 5 LIL Didphragmivalls simis A ide 6 1 3 1 1 Calculation methods iii a aio 6 1 3 1 2 FEMin embedded Walls problem cici meitat td 8 1 3 2 Waterproof concretes at A A A hc SE 9 1 3 3 Fibte teinforced CON aaa dien 10 1 3 4 Sprayed contretes eneee a RRE a E a E A E a OR 11 1 3 5 BONO betiveen Concretes simi iii iii id i iea aaa aaia iaa aie i adioa 13 1 4 GENERAL METHODOLOGY wa occ isiiaiacitmenn atmmemstmitastmaama etstziimsamcienmmnastasmmsmtaiciana aca ens 13 1253 THESIS STRUCTURE ac 15 1 5 1 Chaptets QUICK ii A da dies 17 A A NN 18 CHAPTER 2 EXPERIMENTAL AND NUMERICAL STRUCTURAL ANALYSIS cccccccssssssssssceeecccecesssssssceeeeeeeees 21 2 15 INTRODUCTIONS edit A AA A ticle 22 2 2 EXPERIMENTAL PROGRA Miss Rda daa 23 A ER AR iii pecat cien atscimens a cients vau
133. eseceeeeeneeeneecnsecsseeeseceaeesseenes 2 Fig 2 Bi layer diaphragm walls a general scheme b compound section c simple section and d spraying of an experimental Wall oee eicere rsin e nta e aoea Eia EAE Ea A EEEa 4 Fig 3 Main knowledge areas needed to materialize the bi layer diaphragm vvallS eeoo22o02uo ou 0nueonee 5 Fig 4 Typical ultimate states modes in embedded walls Fuente Eurocode 7 EN 2004a 1 Fig 5 gt Outline of the thesissen eeni e verne niea o EERO E EEE V aaa E 17 Fig 6 Experimental building a site plan b general cross section c detail of bi layer cross section 23 Eig 7 Wall Construetion sequence soenniet e ienaa ene ana eenean temen inde 24 Fig 8 Bi layer walls a Spraying of second layer b finished sprayed surface ooooonocnnncnnocononconncnnncnnnnnnnos 26 Fig 9 Finite Element mesh Of W35 wall 0 ee eececssecseceseceseceseceseceseeeseecaaecaaecaecaeceaeesseceaeseaeesaeesaaesnaees 27 Fig 10 Inclinometer results of W35 wall a Incremental displacements b Check sum values 30 Fig 11 Incremental displacements at the same structural stage and at different times for W35 Reference stage PreElim150 2 33 cc nio a ail win ae eae enda ga seis 31 Fig 12 Comparison of incremental displacement values calculated by the PLAXIS model and experimental values for the representative stages of Wall W35 Reference stage EXC180
134. esta t cnica la impermeabilidad de las pantallas continuas ha sido motivo de debate dando lugar a la aparici n de diversas t cnicas para reparar las filtraciones cuando estas ocurren o prevenir su aparici n Puller 1994 En la Fig 55 se muestran algunos ejemplos t picos de defectos ocurridos en pantallas continuas Fig 55 Defectos usuales en pantallas continuas a Diferencias de posici n entre bataches adyacentes b P rdidas entre juntas durante construcci n c P rdidas que aparecen ya en servicio Una t cnica habitualmente utilizada en la construcci n de pantallas continuas consiste en reparar las zonas defectuosas luego de que se detectan las p rdidas repicando estas reas y restituy ndolas con un mortero expansivo impermeable Dado que las p rdidas suelen aparecer en diversas reas de las pantallas en tiempos distintos incluso luego de finalizadas las obras esta soluci n puede extenderse por per odos indefinidos y requerir varias sesiones de trabajo Por ello esta t cnica es un gran inconveniente tanto para el propietario como para la constructora que se tiene que encargar de las reparaciones Otra soluci n consiste en realizar una capa de mortero u hormig n impermeable en todo el paramento interior de estos muros Al revestirse la totalidad del paramento esta se vuelve una soluci n efectiva pero costosa Wong 1997 Un ejemplo de esta soluci n fue utilizada por Li et al 2008 para estudiar el comportam
135. etected in subsequent core extraction tests due to the intrinsic irregularity of the spraying system Segura Castillo amp Aguado de Cea 2012b After spraying the surface was kept wet for a whole day Bi layer diaphragm walls Experimental and numerical analysis 26 Chapter 2 gt Tole ENS get A ELCAN ms ra b eee Fig 8 Bi layer walls a Spraying of second layer b finished sprayed surface Details of the mixture compositions and surface preparation used for constructing the walls can be found elsewhere Segura Castillo amp Aguado de Cea 2012a Once the bi layer wall had been completed the base slab 70 cm thick and the intermediate slabs 22 cm thick were constructed 2 2 3 Tests and Instrumentation Samples were taken when concreting the walls to characterize the first concrete layer and to determine its compressive strength UNE EN 12390 3 UNE EN 12390 3 2003 During the spraying of the second layer two moulds were filled with the same concrete and the procedure outlined in UNE EN 14488 1 UNE EN 14488 1 2006 was followed cylindrical cores were extracted from the moulds to determine the compressive strength of the second layer UNE EN 12390 3 UNE EN 12390 3 2003 An aluminium inclinometer casing was attached to the steel bar reinforcement cage of each experimental panel The cages were then placed in the excavated area for casting the first concrete layer of the panel Inclinometer measurem
136. evelop cracks which provide a path for water ingress and therefore have certain degree of permeability Brown amp Bruggemann 2002 EN 1538 2010 ICE 2007 Puller 1994 As an example three diaphragm wall joints can be seen in Fig 1 where a the union of panels present a tilt b water leakage appeared during construction and c the water leakage appeared after the structure had been completed Fig 1 Typical defects of diaphragm walls in joints between panels Temporary stop end is the usual system used in joints to connect the panels Brown amp Bruggemann 2002 Even if there is a thin layer of bentonite in the joints which usually have an acceptable degree of waterproofness the deflections in the wall during the excavation process create paths that allow a water flow Ou amp Lee 1987 Several methods have been developed to improve the joints against leakages like water stop joints Puller 1994 or end plates overlapping joint El Razek 1999 Ou amp Lee 1987 Luis Segura Castillo Introduction 3 Notwithstanding the efforts made to achieve waterproofness there is a general consensus among contractors and researchers that there is no effective technique to make joints fully watertight ICE 2007 Puller 1994 Wong 1997 especially if there is a high water table in the exterior of the walls Accordingly ICE guide states that a complementary waterproof system should be added if a good level of watertightness is
137. ez 350 300 250 200 150 100 50 0 50 100 150 200 250 300 350 Bending Moments RN m Fig 26 Ultimate and design moments for the BL60 10 wall 3 6 DISCUSSION Besides the structural solutions that have been presented different systems to deal with leakages on diaphragm walls are compared in this section Two standard systems aiming to ensure a dry inner wall after building a conventional mono layer wall are a Drained cavity DC a second inner wall separated from the diaphragm wall The cavity between them is drained and the water accumulated at the bottom is later pumped out b Waterproof mortar layer WML consists of casting a second layer of waterproof mortar over the inner face of the diaphragm wall This layer is usually about 5 cm width and is cast after the diaphragm wall has been finished without structural function Additionally an optional modification is added to each of the bi layer wall types The Optimized fibres Opt system utilizes the idea introduced at the end of section 5 where the fibres are only placed where strictly necessary i e at depths of between 5 5m and 12 5m using a sprayed concrete without fibres for the rest of the spraying layer The system where fibres are uniformly placed all along the second layer are called Unif to differentiate it from the preceding option The main differences between these systems are summarised in Table 15 It includes the basic material
138. ez se dispone de un lapso de aproximadamente dos horas para aplicar el producto luego del cual el mismo cristaliza reduciendo notablemente la capacidad adherente entre hormigones Si se considera la dificultad de prever con precisi n los tiempos de hormigonado cuando se realiza mediante proyectado se puede concluir que es probable que se encuentren zonas de muy variada resistencia en el v nculo Por otro lado al ser un producto de aplicaci n manual se depende de la experiencia del operario para su colocaci n lo que a ade un factor extra que a ade m s dispersi n a los resultados Se observa un descenso de la resistencia al aumentar la edad Tu amp Kruger 1996 tambi n registraron una ca da de la resistencia luego de 14 d as adjudicando la misma al deterioro que causa en el epoxi el agua que migra desde el hormig n fresco y se acumula gradualmente en la interface unida por el epoxi A1 4 2 Resultados pull off En la Table 23 incluida en el Anexo se presentan los resultados obtenidos para el ensayo pull off Los resultados v lidos para nuestro estudio es decir aquellos con rotura en la interfase entre capas R I en la Table 23 se grafican en las Fig 63 y Fig 64 para la preparaci n superficial por fresado y con Epoxi respectivamente 0 8 0 7 0 6 0 5 0 4 0 3 0 2 0 1 PF 1 10 100 Edad del hormig n de refuerzo dias Resistencia a tracci n c MPa Fig 63 Resistencia a tracci n vs edad
139. f both concretes for the ages at which the shear tests were carried out At these ages the strength of the first phase concrete was calculated on the basis of the results from earlier ages using the concrete maturity formula Neville amp Brooks 2010 assuming Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 85 constant average temperatures throughout the experimental campaign This is a reasonable hypothesis taking into account that the element is of little thickness and is in contact with the ground Table 21 Compressive strength of concrete in both phases Age of 1st phase f Lst phase concrete N mm2 Age of 2nd phase fe 2nd phase concrete days MP walls EP and SP walls concrete days concrete N mm2 7 26 29 30 26 28 33 89 34 82 56 36 83 37 37 87 39 28 38 94 2 30 99 91 39 51 39 05 6 39 22 120 40 94 39 97 35 45 40 These values were calculated according to the concrete maturity equations Neville amp Brooks 2010 5 3 2 Shear test results Among the correctly extracted specimens three were not tested This was due to a fault described in 5 3 2 1 in the first two MP 35 and SP 38 and due to an error in the load press in the third MP 35 The results from the shear test are shown in Table 22 The following information is given for each series name specifying surface preparation type and age at testing number of cores debonded in extraction rejected tests and
140. for the representative stages of Wall W35 Reference stage Exc180 The series of plots in Fig 12 show two that correspond to displacements when the bi layer was still not activated stages Exc380 and Exc630 and two that correspond to displacements after the Spraying stage stages Elim350 2 and Control 2 upon completion of the bi layer wall Similar levels of adjustment were therefore achieved in both the pre and post spraying stages Luis Segura Castillo Experimental and numerical structural analysis 33 The corresponding displacements for VVall VV45 are plotted in Fig 13 It can be seen that notably different levels of adjustments vvere reached in the different stages 1 0 1 0 6 30 s 20 4 2 0 oe Exc630 Control 2 ad Control 2 pe 40 an 4 0 E m E o T 50 N 50 2 E 3 8 F i 3 di a 90 u Qe HO u Q or o 7 0 gt 7 0 pd o o 8 0 8 80 Le 8 v v 9 0 5 90 a 5 Q o 10 0 4 y a 10 0 a 04 0 2 o 0 2 0 4 0 4 0 2 o 0 2 0 4 0 2 0 0 2 04 O Exp Mod A Exp Mod O Exp Mod Incremental Displacements mm Incremental Displacements mm Fig 13 Comparison of incremental displacements calculated with the PLAXIS model and experimentally obtained for the representative stages of Wall W45 Larger qualitative differences between experimental and model data were recorded in this wall although the displacement magnitudes remained within the same range In this case the differen
141. forget how to hate Ozzy Osbourne CHAPTER 5 Evolution of concrete to concrete bond strength at early 4 ages ABSTRACT An innovative structural element typology is proposed referred to as a bi layer diaphragm wall Its two layers are poured and sprayed respectively in two phases the first layer is a standard reinforced concrete diaphragm wall while the second consists of a layer of sprayed concrete with steel fibres which performs a dual waterproofing and structural role Through an experimental campaign our research aims to study the evolution of bond strength between the two concretes at early ages Three preparation techniques were studied milled surfaces milled and epoxy bonded surfaces and saturated milled surfaces The results reveal that the shear strength of milled surfaces follows a typical maturity law regardless of the milling direction In contrast a wide range of results is evident for in situ epoxy bonded surface preparations Keywords shear test fibre sprayed concrete 1 Segura Castillo L amp Aguado de Cea A 2012 Bi layer diaphragm walls Evolution of concrete to concrete bond strength at early ages Construction and Building Materials 31 1 29 37 doi 10 1016 j conbuildmat 2011 12 090 76 Chapter 5 5 1 INTRODUCTION Leakage represents a widespread problem in diaphragm walls built under certain conditions such as enclosures in water bearing ground Ever since the first walls of this type wer
142. go del paramento En la Fig 65 se denot como a al ngulo producido entre el plano promedio del muro y el plano en una posici n espec fica Bi layer diaphragm walls Experimental and numerical analysis 122 Appendix 1 Fig 65 Esquema de irregularidades en la interfase entre capas y superficie de pantallas La segunda capa es formada mediante hormig n proyectado Lo dicho anteriormente respecto a la dificultad en controlar es espesor de la capa de proyectado implica directamente la formaci n de un nuevo ngulo del paramento respecto al plano medio de la pantalla En la Fig 65 se denot como a a dicho ngulo En la Fig 66 se ve una foto de la obra experimental luego de acabo el proyectado Se puede observar claramente el acabado con superficie irregular Para realizar correctamente el ensayo ser a necesario alinear perpendicularmente al plano determinado por a la m quina para realizar la perforaci n parcial del testigo En la experiencia realizada no hemos logrado encontrar una forma de realizar eficientemente esta alineaci n Cabe mencionar que se debieron realizar un conjunto grande de ensayos en un per odo limitado de tiempo disponiendo de poco margen tanto de tiempo como de recursos humanos para solucionar los problemas que surg an en el transcurso de la experimentaci n Fig 66 Foto del acabado final del proyectado en los muros Bi Capa Viendo los resultados obtenidos y considerando los comentarios a
143. he wall All in all you re just another brick in the wall Pink Floyd The Wall CHAPTER 3 Structural and sectional analysis ABSTRACT The bi layer diaphragm wall a new slurry wall type designed to cope with the problem of watertightness is studied in this paper These walls consist of two bonded concrete layers the first a conventional Reinforced Concrete RC diaphragm wall and the second a Sprayed Steel Fibre Reinforced Concrete SFRC The main objective of this paper is to analyze the structural and sectional behaviour of these walls A study in the form of an uncoupled structural section analysis based on various hypothetical cases of bi layer diaphragm walls was performed to fulfil the objective It is concluded that there exists a potential of reduction in the reinforcement of the RC layer through the structural use of the SFRC layer However when the reduction is quantified even though a reduction of between 3 2 and 1 7 in the RC reinforcement is confirmed it appears insufficient to offer a cost effective solution Nonetheless the system becomes a promising solution when particular conditions are taken into account such as basement space limitations Keywords waterproof diaphragm walls fibre concrete sprayed concrete numerical analysis FEM PLAXIS Segura Castillo L Aguado A de la Fuente A amp Josa A 2013 Bi layer diaphragm walls Structural and sectional analysis Journal of Civil Enginee
144. her hand the theories in which the method is based are wide more complex as it is also its application including pre and post processing Furthermore some of the theories and models are still under development Ou 2006 and for every soil model specific soil parameters need usually to be evaluated to have the required input data to acceptably model the soil behaviour Nowadays the more widespread soil models used in FEM studies of diaphragm walls include the well known elastic perfectly plastic Mohr Coulomb MC the modified Cam Clay MCC Roscoe amp Burland 1968 the Hyperbolic Hyp Duncan amp Chang 1970 and the hardening soil HS Schanz Vermeer amp Bonnier 1999 and to a lesser extent more advanced models such as the MIT E3 Whittle 1987 As the accuracy of FEM based models depends to a large extent on the selection of appropriate parameters to represent the constitutive behaviour of soils Khoiri amp Ou 2013 many researchers focused their attention on direct or inverse ways to determine those parameters For example Khoiri amp Ou 2013 used the MC and HS models to predict deformations based on measured data at the first excavation stage also measuring the soil Young s modulus aiming to correlate it with the previous prediction Ou amp Lai 1994 analysed layered sandy and clayey soil using both the Hyp and the MCC models and establishing a procedure to determine the so
145. iento de la fluencia a tracci n del hormig n a edades tempranas En 1990 se public una norma British Standard Code of Practice BS 8102 1990 que defini grados de estanqueidad Luego de esta publicaci n la pr ctica com n en UK consisti en hacer una cavidad drenada c mara bufa con una bomba permanente en un sumidero en el nivel m s bajo Por lo tanto el volumen con que se dise a las bases es reducido por el volumen de la cavidad de drenado el volumen de los muros de revestimiento y el volumen de las tolerancias de construcci n de los muros pantalla El Hussieny 1992 Todo ello conduce a una p rdida de espacio que puede tener una repercusi n negativa con respecto a la funci n requerida del espacio A1 1 2 Marco general del proyecto Este trabajo forma parte de la tesis doctoral del primer autor que se est realizando en la Universidad Polit cnica de Catalunya UPC El objetivo central planteado para esta tesis es el desarrollar evaluando su Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 113 viabilidad modelando num ricamente contrastando experimentalmente e indicando el modo de disefio un elemento estructural nuevo de caracter sticas innovadoras el Muro Pantalla Bi Capa Estos muros presentan una secci n compuesta por dos hormigones a hormig n realizado mediante el sistema de muro pantalla tradicional y b una segunda capa de hormig n proy
146. il parameters and Calvello amp Finno 2004 Hashash Levasseur Osouli Finno amp Malecot 2010 calibrated the soil parameters through back analysis There are many recent examples of studies based on these methods related to the design of diaphragm walls For example improvements of the empirical methods were performed by G T C Kung Juang Hsiao amp Hashash 2007 and Bryson amp Zapata Medina 2012 who proposed new semi empirical methods based on FEM parametrical analysis the construction sequences were analysed and compared by G T C Kung 2009 different soil models MCC two variations of HS MC and undrained soft clay model were analysed and compared by Lim Ou amp Hsieh 2010 under undrained conditions who concluded that all models could predict with a correct parameter selection the wall deflections but only the last one was able to predict surface settlements correctly finally it can be mentioned the Ou Chiou amp Wu 1996 study on the influence of the spatial effects in the wall behaviour It can be concluded that in the last years the FEM method using the abovementioned soil models has been frequently and increasingly being used to address wide different aspects of the diaphragm wall technique In Luis Segura Castillo Introduction 9 the majority of studies the focus is put both in the displacements of the wall and of the adjacent soil mainly due to the importance they have to avoid da
147. ile the second consists of a layer of sprayed concrete with steel fibers which performs a dual waterproofing and structural role The bond between both concretes plays an important role in the performance of the structure If bond strength is sufficiently high the structure behaves monolithically Through an experimental campaign our research aims to study the evolution of bond strength between the two concretes at early ages 2 6 and 35 days Two preparation techniques were studied milled surfaces and milled and epoxy bonded surfaces The bond strength was assessed through shear and pull off tests The results reveal that the shear strength of milled surfaces follows a typical maturity law In contrast a wide range of results is evident for in situ epoxy bonded surface preparations Pull off tests show in every case a wide range of results Keywords diaphragm wall bi layer waterproof bond shear pull off concrete fibers 3 Segura Castillo L amp Aguado de Cea A 2012 Bi layer diaphragm walls Early ages concrete to concrete bond strength assessed through shear and pull off tests in spanish In XXXV Jornadas Sudamericanas de Ingenier a Estructural R o de Janeiro ASAEE 12 Appendix 1 A1 1 INTRODUCCI N A1 1 1 Impermeabilizaci n en pantallas continuas La aparici n de filtraciones de agua es un problema habitual en las pantallas continuas realizadas en terrenos con un nivel fre tico elevado Desde la aparici n de
148. in the supports The test was performed using a hydraulic press with displacement control Koi pp L 2 94 mm L 2 94 mm Fig 43 LCB shear test a device sketch b test configuration The shear stress which appears in the bonded interface is calculated according to the following formula t P 2 S 7 where t is the shear stress MPa P is the maximum load at failure N and S is the area of the cross section of the specimen mm Prior to the extraction of the cores a mark was made on the wall indicating its vertical direction which coincided with the milling direction Using this mark as a reference the cores were oriented to perform shear stress in a direction perpendicular to milling VM parallel to milling HM and in a direction offset by 45 in relation to the previous ones OM Fig 44 shows a sketch of these positions The core in Fig 43b has a mark in the horizontal position i e the stress runs perpendicular to the direction of milling Milling direction mark WM HM OM Fig 44 Core positions for shear test 5 3 RESULTS AND DISCUSSION 5 3 1 Mechanical characterization results Table 21 shows the evolution of strength as regards the age of both concrete phases As can be seen the strength values obtained for all concretes were above the expected design values presented in section 5 2 1 The last three lines of the table show the compressive strength o
149. ing sequence is a crucial parameter to be able to take advantage of the SFRC collaboration and specific indications are described Good concrete to concrete bond strength was obtained for the extracted cores The average shear strength value measured for each age 2 6 and 35 days was always above 1 0 MPa for the different cases Beyond the local test performed a monolithic behaviour was observed at element level in the experimental walls A similar final material consumption was observed between the BL walls and the combined consideration of a ML wall and an external waterproof system The consideration of the technology cost entails a higher construction cost for the BL technique However it is still an interesting option under particular circumstances like space limitations or if continuous maintenance costs want to be avoided in the future In general terms it can be said that the research herein presented lay the foundation for the development of the bi layer diaphragm wall technique which is a promising solution for the leakage problem of diaphragm walls Nonetheless more studies are needed to be able to fully use these types of walls as a standard Bi layer diaphragm walls Experimental and numerical analysis iv Summary technique e g a detailed cost study and sustainability analysis debonding risk waterproofing capability and above all more full scale experimental cases Luis Segura Castillo Summary v RESUMEN Un probl
150. ing the complete systems including both the structural and the waterproof system the Opt systems are more efficient than the Unif systems and will therefore be used in all subsequent comparisons The MLso DC system is nowadays one of the more commonly used for waterproofing the wall surface The material required for the drained cavity leaf wall and extraction pump is assumed to be relatively low The main drawbacks of this system are a Reduced interior space crucial in the basements of buildings designed for underground parking and other economic activities b Need to activate a pump as excess water accumulates and c It hides the source and the extent of the leakages or any other possible structural problem that the walls may have Puller 1994 Bi layer diaphragm walls Experimental and numerical analysis 54 Chapter 3 If the use of materials of the ML WML system is compared with the BLsos io Opt the latter registers an increase of 5 0 in concrete use and 2 3 in steel Considering that the thickness of the BL system is also larger the BL alternative is not favourable in this case If the ML WML system is compared with the BLss 0 Opt it should first of all be noted that both have the same final thickness and are also the slenderest of all the waterproof systems under study Regarding the materials it can be seen that a reduction of 3 0 in the amount of concrete the only material related value
151. ition of the measures is plotted 2 0 2 0 a 15 a E E 10 Ca E E DT a v Zos l NY 7 amp 10 r O a 0 0 0 de RARAS eitti E os Ear LF 3 05 7 o 5 S o 10 A o SES 5 15 r r T r r r r r 0 5 r r 1 1 1 00 10 20 30 40 50 60 70 80 90 100 00 10 20 30 40 50 60 70 80 90 100 Depth m Depth m OCast Exc180 A Anc150 O Exc380 OAnc350 Excint 1 A Exc630 O PreElim350 m Elim350 1 Elim350 2 a PreElim150 e Elim150 1 mElim150 2 Elim150 3 a Control 1 Control 2 Fig 10 Inclinometer results of W35 wall a Incremental displacements b Check sum values Luis Segura Castillo Experimental and numerical structural analysis 31 The plot of the Check sum values is shown in Fig 10b Linear trend lines of the values of the offset for every stage are also shown in Fig 10b Systematic errors at these stages can be identified by observing these trend lines The check sum is usually equal to twice the zero offset bias of the transducer and ideally should remain constant for all depth intervals in a given data set Check sum may vary randomly about a mean value Small variations do not usually indicate a problem Dunnicliff 1993 The area where 13 out of 16 trend lines are concentrated is shadowed in grey and the three trend lines clearly out of this area are individually plotted Small variations are due to the experimental error However the lines further away f
152. ixtures for Waterproofing Construction p 41 Retrieved from http www archsd gov hk media 11756 c315 pdf Clough G W amp O Rourke T D 1990 Construction induced movements of insitu walls In Design and performance of earth retaining structures GSP 25 pp 439 470 ASCE CNR 2006 CNR DT 204 2006 Guide for the Design and Construction of Fiber Reinforced Concrete Structures Costa P A Borges J L amp Fernandes M M 2007 Analysis of A Braced Excavation In Soft Soils Considering The Consolidation Effect Geotechnical and Geological Engineering 25 6 617 629 doi 10 1007 s10706 007 9134 7 Luis Segura Castillo References 103 CPH 2008 EHE 08 Instrucci n del Hormig n Estructural in Spanish CYPE Ingenieros 2011 Software for Architecture Engineering amp Construction Retrieved from http www cype es DBV 2001 Guide to Good Practice Steel Fibre Concrete German Society for Concrete and Construction Technology Berlin De la Fuente A Aguado de Cea A amp Molins C 2008 Numerical model for the nonlinear analysis of precast and sequentially constructed sections in Spanish Hormig n amp Acero 57 247 69 87 De la Fuente A Aguado de Cea A Molins C amp Armengou J 2012 Numerical model for the analysis up to failure of precast concrete sections Computers amp Structures 106 107 105 114 doi 10 1016 j compstruc 2012 04 007 De la Fuente A Escar
153. iz R C de Figueiredo A D Molins C amp Aguado de Cea A 2012 A new design method for steel fibre reinforced concrete pipes Construction and Building Materials 30 547 553 doi 10 1016 j conbuildmat 2011 12 015 Delatte Jr N J Wade D M amp Fowler D W 2000 Laboratory and field testing of concrete bond development for expedited bonded concrete overlays Materials Journal 97 3 272 280 Retrieved from http www concrete org PUBS JOURNALS AbstractDetails asp ID 4622 Delatte N J Williamson M S amp Fowler D W 2000 Bond strength development with maturity of high early strength bonded concrete overlays ACI Materials Journal 97 2 201 207 Retrieved from http www concrete org PUBS JOURNALS OLJDetails asp Home MJ amp ID 824 Delattre L 2001 A century of design methods for retaining walls The French point of view 33 52 Di Prisco M Plizzari G amp Vandewalle L 2009 Fibre reinforced concrete new design perspectives Materials and Structures 42 9 1261 1281 doi 10 1617 s11527 009 9529 4 Duncan J amp Chang C 1970 Nonlinear analysis of stress and strain in soils Journal of the Soil Mechanics and Foundations Division 96 5 1629 1653 Retrieved from http cedb asce org cgi VV VV VVdisplay cgi217188 Dunnicliff J 1993 Geotechnical instrumentation for monitoring field performance New York John Wiley amp Sons Durmisevic S 1999 The future of the underg
154. ken over one structural stage and discarding the readings taken just after anchorage installations 2 5 MODEL VS EXPERIMENTAL COMPARISON 2 5 1 Model adjustment The experimental incremental displacements are plotted with error bars in Fig 12 through the representative stages of Wall W35 The error bars indicate the standard deviation of the Check Sum of each inclinometer The corresponding values obtained by the model are plotted with a continuous line in the same figure Worse adjustment was registered for Wall W35 at stage Exc380 at depths of below 3 0 m The auxiliary anchorage was positioned at a depth of 3 5 m on this panel after partial excavation in the area surrounding the panel and not when the general site excavation had reached the necessary depth A dragging effect from the neighbouring panels might have been the cause of these differences between the model and the experimental data as soil still surrounding the side panels might have resisted any lateral displacement towards the excavated area caused by the partial excavation in the experimental panel Exc380 Exc630 Elim350 2 Control 2 40 Depth m 0 4 0 2 0 0 2 0 4 0 6 0 4 0 2 0 0 2 04 0 8 0 6 0 4 0 2 0 0 2 0 4 0 4 02 o 0 2 0 4 O Exp Mod A Exp Mod Exp Mod O Exp Mod Incremental Displacements mm Fig 12 Comparison of incremental displacement values calculated by the PLAXIS model and experimental values
155. lable from the Spanish Ministry of Education and Science through the Research Project BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras Bi layer diaphragm walls Experimental and numerical analysis 94 Luis Segura Castillo Chapter 5 CHAPTER 6 CONCLUSIONS AND FUTURE PERSPECTIVES 6 1 INTRODUCTION Underground space use is becoming vital to the developing of modern cities The diaphragm wall technique which causes a limited influence on existing infrastructure and also reduced interruptions to the daily life of the city during construction is a viable solution to the construction of underground structures in a city scenario A common problem associated with diaphragm walls is that they are frequently not fully watertight The aim of the research was to develop an innovative type of slurry wall the bi layer diaphragm wall which will offer a new solution to the waterproof problem in diaphragm walls The bi layer walls are made of two bonded concrete layers the first is a conventional reinforced concrete diaphragm wall and the second is made spraying steel fibre reinforced concrete with a waterproof additive over the first layer A full scale experiment where two bi layer walls of different cross sections were constructed was performed and studied followed by theoretical analysis to corroborate the advantages provided by the solution The study was centred in three key aspects of this type of walls the
156. labs slabs Anc Out Anc Out Excavation base Anchorage position or Spraying base according to the respective stage The soil extraction process for the BLso 9 and BLs5 19 bi layers walls is sub divided into shorter stages In addition after each partial excavation stage the SFRC layer is Sprayed Spray from the last sprayed level to the lower excavated level changing the cross section from the Simple Section SS see Fig 21b to the Compound Section CS see Fig 21c in the sprayed stretch 3 3 2 Material and model characteristics The numerical model was calculated on the commercial geotechnical finite element software package PLAXIS Brinkgreve 2002 The FEM mesh used is shown in Fig 22 Horizontal fixity was imposed for the vertical boundaries as well as both horizontal and vertical fixities for the bottom boundary as shown in the same figure A fine global coarseness was taken for the general mesh automatically defined by the program and refined in the vicinity of the wall A model with a more refined mesh verified that the element size had no significant effects on the analytical results Besides no external loads were considered in the model Plate structural elements linear elastic were used to model the diaphragm walls which were considered wished in place Bryson Zapata Medina 2012 A compressive strength fe 30 MPa a Poisson ratio v Bi layer diaphragm walls Experimental and numerical analysis 46 Ch
157. ll W35 The same process was followed for Wall W45 although its description is omitted here The values of the incremental displacements Ij are shown in Fig 10a Individual errors are abnormal values recorded at a specific stage and at a specific depth for example the four measures that are circled in Fig 10a that are clearly beyond the normal range of displacements of the wall Moreover these points lie outside the normal range of recorded values for the offset These disproportionate displacements may be produced by local deformations induced by the anchors as they occurred at stages when the anchors were activated and at points near the positions of the anchors Systematic error can be of two types abnormal values repeatedly registered at various depths throughout the same stage or at different stages at the same depth Two depths 6 5 m amp 9 5 m are circled in Fig 10a The readings at a depth of 9 5 m were taken from the bottom of the inclinometer tube As it was the first measurement in each series this reading could have been taken while the inclinometer sensor was still not properly stabilized The inconsistencies registered at depth 6 5 m might be due to an imperfection in the inclinometer tube At this depth there is a coupling between two sections of the inclinometer casing A slight break could be seen in the connection in a photo taken during the casting of the wall Furthermore the break can be seen if the absolute pos
158. lona The campaign was focused on three aspects a assess viability of the general solution b structural behaviour of the walls and c bond between layers For the structural experimentation displacements were measured by means of inclinometers and invar tape strain gauges were placed in the reinforcement bars and load cells at the anchorage points of two instrumented panels Despite the different type of structural measurements due to different problems Bi layer diaphragm walls Experimental and numerical analysis 14 M2 M3 M4 MS M6 M7 M8 MO M10 Chapter 1 related to the on site experimentation load cell broken need to change the reference point of the invar tape measurement strain gages measures missed the analysis was finally exclusively based on the inclinometers results which had complete and reliable data For the bond between layers experimentation cores from the bi layers walls were extracted at different ages and taken to the lab to perform shear tests A direct shear test feasible to be used on extracted cores was adopted Some adaptations were performed to adjust the test originally designed for bituminous materials to be used with concrete cores Using the developed test an experimental campaign evaluating the evolution at early ages of the bond strength was performed A structural model for the walls behaviour was developed Two types of soil structure interaction mod
159. lysis 62 Chapter 4 position where the supports are positioned The BU construction sequences for the 2u case are equivalent to those shown here with the difference that in the slabs stage the slabs that correspond to the 2u final design are built see Fig 28a The construction sequence TD for the 2u case is represented in Fig 29b Four additional stages must also be considered for the TD sequence of the 4u case two intermediate excavations 1 75 m below the slabs 4 75 m and 10 75 m deep necessary for the additional intermediate slabs at depths of 3 00 m and 9 00 m o Bottom up BUa BUs for 4u case Wall const Exc 12 50 5 5 N N Anchorage or Strut depending on each case b Top Down TD for 2u case 7 a Q 6 Fig 29 Construction sequences a Bottom Up BU b Top Down TD Fig 30 shows examples of the different depths of sprayed concrete layer The temporal supports and slabs which should be considered according to each case are omitted from the figure The SFRC is sprayed when the depth of the excavation is 25 cm below the depth indicated for each spraying section in case of any possible excavation irregularities The three alternatives for the DS parameter indicated in Fig 30 were only studied in cases that involved two spraying stages 2S When the excavation reaches the final depth the last section of the second layer is
160. mage to existing buildings Despite the difficulty to calibrate soil parameters the FEM method seems to give good results regarding wall deflections with all the above mentioned soil models therefore it was chosen for the design of the cases analysed in this thesis Among the soil models the HS soil model was chosen mainly for two practical reasons Firstly it has been previously used in other diaphragm wall studies were sand had to be modelled Therefore there are good quality documented cases with soil parameters Secondly it was developed associated with PLAXIS the FEM program available to carry out the calculations 1 3 2 Waterproof concrete Concrete is a porous material It is possible to water to penetrate both through its porous and for its microcracks due to capillary absorption or due to hydrostatic pressure Despite low water cement ratio concrete properly produced has generally good durability and low permeability no concrete can be considered absolutely waterproof ACI Committee 212 2010 However it may be possible to reduce the permeability of the concrete of the second layer of the proposed solution to be considered sufficiently waterproof for the desired application In this sense there are a range of products called permeability reducing admixtures PRAs with variances in performance capable of reducing the concrete permeability ACI Committee 212 2010 Chan Ho amp Chan 1999 Ramachandran 1995 can be named as g
161. mbr la semilla de la curiosidad por la vida me ense a dudar a mirar con ojo cr tico la vida Con Juancito exploramos y explotamos en discusiones metaf sicas existenciales Ese hurac n de ideas sentimientos y pasiones que eran el e hicieron temblar mis cimientos Pocho es un faro de luz que sigue iluminando marcando el rumbo al que quiero ir en el que la calidad humana est por encima de todo A los tres Gracias Luis Segura Castillo Summary iii SUMMARY Leakage is a widespread problem associated with the construction of diaphragm walls whenever they are erected in water bearing ground The aim of the present research is to develop a new type of slurry wall the bi layer diaphragm wall BL which main objective is to tackle the aforementioned problem The method to construct it is based on an existing solution casting a second waterproof concrete layer against the diaphragm walls In the BL technique the second layer is made of steel fibre reinforced concrete SFRC sprayed over the conventional diaphragm wall called Mono Layer diaphragm wall ML in this thesis including a waterproof admixture The central idea is to maximize the functional attributes of the second layer allowing it to play a structural role in addition to the waterproofing function The proposed methodology is based on a combination of experimental works and numerical tools A design method for the BL walls which is based on an uncoupled structural section
162. ment around depth 10 0 m even decrease In contrast the final configuration in the 2u case see Fig 35b is not as stiff 3 slabs as in the 4u configuration Therefore a larger redistribution of forces occurs and particularly an increase in the maximum bending moments after the final excavation see Fig 35b between depths 7 0 m and 11 0 m This increase can be resisted with the additional strength given by the SFRC layer i e use the additional strength given by the second layer even if it is sprayed after the excavation is finished Bi layer diaphragm walls Experimental and numerical analysis 70 Chapter 4 Fig 36 shovvs the horizontal displacements for the corresponding cases shovvn in Fig 35 It also includes the displacements in the respective mono layer cases ML 60 0 BUs 4u 0S M for the 4u case and ML 60 0 BUs 2u 0S M for the 2u case A e See ee eee Es di CAN DOU uN FE O oa pus o 9 5 Depth m fd n Pp a NU NL 2u BL NU NL 2u ML NU NL 4u BL NU NL 4u ML IS ou 11 10 9 8 7 6 5 4 3 2 1 0 Horizontal Displacements mm Fig 36 Displacements for different final structure configurations As is reasonable to assume the maximum displacement in the 2u case is larger 0 4 mm than in the 4u case in accordance with the different stiffnesses of both final configurations Once again there are no appreciable differen
163. mental and numerical structural analysis ABSTRACT The bi layer diaphragm wall a new type of wall consists of two concrete layers the first of which is poured and the second sprayed in different construction stages A major aim of the research conducted is to maximize the functional attributes of the second layer enhancing both structural performance and watertightness The central objective of this study is to corroborate the structural behaviour of these walls in experimental and numerical terms lt follows a three step methodology a full scale experimental campaign development of a Finite Element Model FEM capable of predicting the structural behaviour of the wall and assessment of the second layer contribution The experimental campaign confirmed the viability of the constructive solution and the FEM model accurately reflected the experimental data A comparison between the bi layer wall and other single layer walls showed that the contribution of the second layer permitted reductions in first layer reinforcement adding to its various other functional advantages Keywords Fibre concrete Sprayed concrete Numerical analysis FEM PLAXIS Watertightness Segura Castillo L Aguado A amp Josa A 2013 Bi layer diaphragm walls Experimental and numerical structural analysis Engineering Structures 56 154 164 doi 10 1016 j engstruct 2013 04 018 22 Chapter 2 2 1 INTRODUCTION Large cities are encouraged to make effici
164. mm o 0 0 0 1 0 2 T r x 0 0 10 20 30 40 50 60 70 80 90 10 0 Depth m Fig 11 Incremental displacements at the same structural stage and at different times for W35 Reference stage PreElim150 Two readings were taken on the day the anchorages were installed one before the installation and one immediately after it The soil could only develop instant deformations between these stages As the excavation resumed after installation of the anchors the effects of any subsequent soil deformation were combined with the excavation effects A clear reading of the situation after anchorage installation could not therefore be taken so the readings taken just after these stages could not be used in the structural analysis Bi layer diaphragm walls Experimental and numerical analysis 32 Chapter 2 Only very slight incremental displacements were observed at some stages particularly in Wall VV45 that have greater flexural stiffness Therefore the results were qualitatively evaluated in the stages with small displacements i e Wall W45 while the stages with large displacements i e Wall W35 were used for the quantitative adjustment of the model parameters The most representative stages on which to perform the structural analysis underlined in Table 5 were selected on the basis of the above mentioned criteria i e Elimination of individual and systematic measurement errors selecting the last reading when several readings were ta
165. model is proposed The method is later used to carry out different comparisons with ML walls and an exhaustive parametric analysis of the construction processes involved in the walls construction The experimental campaign comprised test at two levels At element level the structural response of walls built in a real building located in Barcelona was studied and at section level the bond strength between concretes of cores extracted from the abovementioned walls was measured The model at structural level which is based on a finite element model was contrasted with the results obtained in the experimental walls The sectional analysis is taken from the specialized literature With the complete structural section model the BL walls are analysed The study shows that the main flexural resistance is provided by the first layer the conventional diaphragm wall providing the SFRC layer a secondary flexural resistance For the geometrical ranges of the elements considered in the thesis 35 cm to 60 cm width first layer and 10 cm width second layer the increase in the cross section ultimate bending resistance when it is strengthened by the SFRC layer is between 8 and 15 This increase allows a reduction in the steel reinforcement of the first layer up to 7 0 of the total flexural reinforcement and to some extent it also collaborates with a displacement reduction reducing up to 7 3 of the maximum displacements It was also found that the spray
166. n Ita siempre al firme para ver o hacer rock A la fl a en Uruguay mam Lauri Valen Moni Marga el piti Daisy que se la re jug con el t tulo Walter Shirly Carla Bruno Martina y la poripocha Y la fl a en Europa Martina y Francis Maia Marcelo Sara y Violeta Mar a y Jorge que me bancaron en m s de un viaje A las barras incondicionales de Uruguay que en alg n u otro momento estuvieron para alentar Toda la cadena de Letizia que siempre tir pa arriba Silvina Isabel Pedro Gime Regi Vale y Sebas Mati y Pame Henry Piti Antonio Sabri Fito Ichu Damian Pepe Geral Naty Nadia Javi rock y Pilar y la gente m s civilizada Mari y Sebas Juampi Sofi y Andres Niky Vero y Fede Joaqu n Ceci y Crufi Gago Mart n y Bi layer diaphragm walls Experimental and numerical analysis il Acknowledgments Dani y Maxi y Gabi Gracias Ani por tu tiempo por aqui Gracias a toda la familia Coscia Omar tus correos fueron un cable directo con el paisito al igual que las novedades y fotos y hasta alg n libro que me tiraba Mario de vez en cuando Thanks to Simon and Chris for the opportunity to live the great academic personal and Olympic experience in Loughborough The two big lofbra guys Dave A M amp James A K you make a wonderful team hope to see you soon Pratesinha bonina thanks for the long nights of wine laugh and crazy talks And F please remember Por ultimo quiero destacar a person
167. n Fig 34b Considering both aspects the design of the bi layer walls is in these cases sensitive to variations in the parameter under study The strength increase provided by the SFRC layer can only be used in the intermediate situation case B red envelope in Fig 34 while it is of no use in the extreme situations case A blue envelope and case C green envelope in Fig 34 So the possibility of taking full advantage of the bi layer walls depends on a correct selection of the spraying sequence This is a negative factor of the technique complicating both the design and the construction of the wall as 1t entails careful control over excavation depths and spraying sequences on the building site Thirdly it can be seen that the value of the maximum moment differs in the three cases As seen in the previous sections the earlier the SFRC is sprayed the larger the value of the bending moment where a maximum difference of 11 2kN m is observed 3 8 for case C as a percentage of case A Luis Segura Castillo Parametric study of construction processes 69 Finally it is vvorth mentioning that although the depth of the sprayed concrete layer vvas analysed tovvards the end of the excavation the results can be extrapolated to an intermediate excavation situation at the maximum local moment between depths of 4 0 m and 6 0 m Besides as described in section 4 3 1 in a similar way the deformations associated with the three cas
168. nal Conference on Soil Mechanics and Foundation Engineering pp 309 312 Kyoto Japan PLAXIS 2D 2010a Reference Manual Retrieved from http www plaxis nl files files 2D2010 2 Reference_02 pdf PLAXIS 2D 2010b Tutorial Manual Retrieved December 01 2011 from http www plaxis nl files files 2D2010 2 Reference_02 pdf Potts D M amp Zdravkovi L 1999 Finite Element Analysis in Geotechnical Engineering Theory London Telford Puller M 1994 The waterproofness of structural diaphragm walls Proceedings of the ICE Geotechnical Engineering 107 1 47 57 doi 10 1680 igeng 1994 25720 Ramachandran V S 1995 Concrete Admixtures Handbook Properties Science and Technology 2nd edition p 1180 Noyes Publications Ray I Davalos J amp Luo S 2005 Interface evaluations of overlay concrete bi layer composites by a direct shear test method Cement and Concrete Composites 27 3 339 347 doi 10 1016 j cemconcomp 2004 02 048 RILEM TC 162 TDF 2003 o s Design Method Materials and Structures 36 8 560 567 doi http dx doi org 10 1007 BF02480834 PB Rodriguez Li an C 1995 Pantallas para excavaciones profundas Construcci n y C lculo p 160 Sevilla Escuela T cnica Superior de Arquitectura de Sevilla R nk K Ritola J amp Rauhala K 1998 Underground space in land use planning Tunnelling and Underground Space Technology 13 1 39 49 doi 10 1016 50886 7798 98 00029 7
169. nd Structures 28 5 249 259 Retrieved from http link springer com article 10 1007 BF02473259 Austin Simon Robins P amp Pan Y 1999 Shear bond testing of concrete repairs Cement and concrete research 29 1067 1076 Retrieved from http linkinghub elsevier com retrieve p11 S0008884699000885 Bentur A amp Mindess S 2007 Fibre Reinforced Cementitious Composites p 625 Taylor amp Francis Group Blanco A Pujadas P de la Fuente A Cavalaro S amp Aguado A 2013 Application of constitutive models in European codes to RC FRC Construction and Building Materials 40 246 259 doi 10 1016 j conbuildmat 2012 09 096 Bobylev N 2006 Strategic environmental assessment of urban underground infrastructure development policies Tunnelling and Underground Space Technology 21 3 4 469 doi 10 1016 tust 2005 12 106 Bi layer diaphragm walls Experimental and numerical analysis 102 References Bonaldo E Barros J amp Lourenco P 2005 Bond characterization between concrete substrate and repairing SFRC using pull off testing International Journal of Adhesion and Adhesives 25 6 463 474 doi 10 1016 j ijadhadh 2005 01 002 Bose S K amp Som N N 1998 Parametric study of a braced cut by finite element method Computers and Geotechnics 22 2 91 107 doi 10 1016 50266 352X 97 00033 5 Brinkgreve R 2002 PLAXIS 2D Version 8 A A Balkema Publishers Lisse The Netherlands p 18
170. nd strength M3 The objective was to establish a simpler test to be used as a routine bond test in this type of walls The results of the paper 4 are summarized in this conference paper and correlated with the pull off results As it was not possible to extract conclusive results from this correlation it was decided to present this paper as an appendix separated of the thesis main body 1 6 RESERCH FRAMEWORK Since the incorporation to the research group headed by professor Antonio Aguado the candidate has participated in different research projects The two main ones are directly related to the topic developed in this thesis In the first place the candidate is part of the research team working in the Spanish Ministry of Science and Innovation MICINN project CONSFIB reference BIA 1010 17478 Construction processes Luis Segura Castillo Introduction 19 by means of fibre reinforced concretes Secondly the candidate participated in the PERMASTOP project CTT 8062 This was an enterprise university project which aim was to develop the bi layer diaphragm walls Bi layer diaphragm walls Experimental and numerical analysis 20 Luis Segura Castillo Chapter 1 Lo que hagamos debe tener algo que podriamos llamar econom a c smica estar de acuerdo con el orden profundo del mundo y s lo entonces podr tener esa autoridad que tanto nos sorprende frente a las grandes obras del pasado Eladio Dieste CHAPTER 2 Experi
171. needed ICE 2007 With this porpoise several techniques have been developed to prevent or repair the existence of leaks Puller 1994 A usual technique is to repair the affected areas as they are detected injecting a chemical grout into the problematic cracks and joints or directly chipping and restoring the element with a more waterproof mortar also called grouting Wong 1997 In a similar way a cement or chemical grout can be injected in the soil behind the wall in the areas where leakage is observed Puller 1994 El Razek 1999 reported having successfully used this solution in a diaphragm wall project in Alexandria However leakage usually appears only over lengthy periods and at different times and areas of a wall see Fig 1c even becoming worse with time Wong 1997 which may result in extensive repair works over indefinite periods that require several sessions causing problems for both owners and contractors Another solution consists of casting an additional layer of mortar or concrete on the interior face of the walls also called tanking which is also a common way to make walls of bored piles watertight Wong 1997 As it can be seen in Fig 1 it is not strange that deviations from true verticality occur in the panels as they are constructed A favourable effect of this technique is that the second layer evens the surface when irregularities caused by panel deviations are detected Few publications were found reporti
172. neeeaeeeaeeenaes 47 Fig 24 Bending moments a Envelopes for the three wall types b Representative stages and envelopes for the BL60 10 wall type c envelope areas for all three wall types oonconccnnnnnnconoccnoccconacnno 48 Fig 25 Diagrams of moments at stages prior to the second anchor installation ooonnncnnnnnncnnncnnncnnncnnncnnnnos 49 Fig 26 Ultimate and design moments for the BL60 10 wall oo ee ee eeeeeseeeeseceseceseceaeceseceseesseeeseeenaes 52 Fig 27 Bi layer diaphragm walls a general scheme b compound cross section c simple cross section and d spraying of an experimental Wall oo cece ceseceseceseeseesseeeseeeaeecsaecsaecaeeseeeseeseeeeeaeeeneeeaaeenaes 58 Fig 28 Final construction design a 2 levels 2u b 4 levels 4u oo cece cecceeneceececeteeeenaeceeaaeceeneeeeas 61 Fig 29 Construction sequences a Bottom Up BU b Top Down TD eee ceseceneceeeeeeeeeeeeenee 62 Fig 30 Spraying discretization considered coooconccnoccnocnnocnnonnconoconaconoconocnnncnnnnnnn cana c nn nono nono naco nn cnn cnn na nan rnn conan 62 Fig 31 Example of bi layer wall bending moment envelopes ooooconocccocccoccconcnonnconncno nono noconoconocanccnnccan crac 64 Fig 32 Bending moment envelope for different spraying discretizations a 1 spraying stage b 2 spraying stages C 4 Spiayime staseSt esti eae bas e n ee e E aeee success vareaetsdepurgaiods dot decrets
173. nelling and Underground Space Technology 7 4 383 391 doi 10 1016 0886 7798 92 90068 S Gaba A Simpson B Powrie W amp Beadman D 2003 CIRIA C580 Embedded retaining walls guidance for economic design Proceedings of the Institution of Civil Engineers Geotechnical engineering Vol 156 p 390 London CIRIA Retrieved from http cat inist fr aModele afficheN amp amp cpsidt 17373224 Gallovich Sarzalejo A Rossi B Perri G Winterberg R amp Perri Aristeguieta R E 2005 Fibras como elemento estructural para el refuerzo del hormig n Manual T cnico In Spanish p 251 Maccaferri do Brasil Ltd Galobardes I 2013 Characterization and control of wet mix sprayed concrete with accelerators Doctoral thesis Univesidad Polit cnica de Catalunya Garcia Vicente T Agull Fit L Aguado de Cea A amp Rodriguez Barboza J U 2001 Propuesta metodol gica para dosificaci n del hormig n proyectado HORMIGON Y ACERO 220 43 56 Retrieved from http dialnet unirioja es servlet articulo codigo 292793 Goodier C I 2000 Wet Process Sprayed Mortar and Concrete for Repair Doctoral dissertation Loughborough University Hashash Y M A Levasseur S Osouli A Finno R amp Malecot Y 2010 Comparison of two inverse analysis techniques for learning deep excavation response Computers and Geotechnics 37 3 323 333 doi 10 1016 j compgeo 2009 11 005 Luis Segura Castillo Refe
174. nes indicate the moments of the stage and light grey lines indicate the moments of the previous stages The envelopes of moments already shown in Fig 24 are obtained when the following process explained below is applied to all the stages Spray 2a Exc 2b Spray 2b l Dj aa d Exc 1 Anclaje 1 Exc 2a Exc 2b mss MS Depth m 80 60 40 20 0 20 80 60 40 20 0 20 Bending Moments kN m Fig 25 Diagrams of moments at stages prior to the second anchor installation Bi layer diaphragm walls Experimental and numerical analysis 50 Chapter 3 It can be seen that the bending moments of the excavation stage Fig 25a are identical to those of the stage where the SFRC is sprayed Fig 25b This is because immediately after spraying all the SFRC layer adds is its own weight which is considered in the model by updating the weight of the beam element This implies a small change in the normal stresses of the element and an insignificant change in the bending moments The external loads on the wall remain constant until a new excavation stage takes place The differential time dependent strains between RC and SFRC layers are left out of the model bearing in mind that the RC layer as a slurry wall has a high confined water content before excavation and that the sprayed SFRC layer with a waterproofing additive has a greater capability of withholding moisture After the spraying
175. ng reinforcement The displacements are almost completely governed by the thickness of the first layer and a minor reduction less than 7 3 is obtained when the second layer is included Keywords fiber concrete sprayed concrete numerical analysis FEM PLAXIS watertightness gt Segura Castillo L Josa A amp Aguado A n d Bi layer diaphragm walls Parametric study of construction processes Engineering Structures Submitted 58 Chapter 4 4 1 INTRODUCTION Diaphragm walls are hardly ever fully watertight as there is generally a degree of permeability between their panel joints Brown amp Bruggemann 2002 Hence some techniques have been developed to deal with the leakage problem in diaphragm walls built in water bearing ground Puller 1994 The bi layer diaphragm wall Segura Castillo Aguado amp Josa 2013 is a new type of slurry wall mainly designed to counter leakage The waterproofing system added in the course of internal site excavations assumes a structural function as an integral part of the wall structure A generic solution and part of the construction of the first experimental walls of this type Segura Castillo et al Segura Castillo Aguado amp Josa 2013 can be seen in Fig 27 These walls consist of two bonded concrete layers poured and then sprayed in separate stages The first is a conventional Reinforced Concrete RC diaphragm wall which forms the simple cross section see Fig 27c
176. ng this solution For example Li Ju Han amp Zhou 2008 used it as a way to study tensile creep in concrete Two arranges for the connection between linings that reflect different possibilities to be used in a real underground structure were studied a continuous design which connect both layers with extruding reinforcements and b sliding design which put an impermeable sheet in between both concrete layers Meanwhile Sherif amp Kudsi 1975 performed a risk analysis on a double wall aiming to quantify the cracking probability which would lead to leaks Since the publication of codes that define three grades of waterproofing protection for underground structures e g BS 8102 2009 or its previous versions it became common practice in the United Kingdom and spread to other countries like USA or Germany to deal with the waterproof problem by constructing an inner wall separated by a cavity Puller 1994 In this system the water is directed to an inferior deposit and pumped out from there The solution is referred as false wall Wong 1997 or drained cavity BS 8102 2009 Nowadays drained cavities are extensively used However this solution loses significant volume because of the construction tolerances and the cavity In some cases the extra space required may have a crucial influence in the final project value For example if an inner leaf 15 cm wide separated by a cavity 10 cm wide is considered neglecting construction
177. nnn cra crac cra 53 Table 16 Parameters and alternatives for each case oooonocnnocnnoconococonoconannncnannnnn nono ncnn nono nc nn ncnnn nac cnn cra cra cra 61 Table 17 General results for all Cds iii a e a 63 Table 18 General results for the additional CasesS oooconnccninnnnocononononcnnnconnnonn nono cono nono nonn cc no cnn ncnn crac nannncnn nina nos 75 Table 19 Concret dosages ene a tooo DICE saan tk etat cot iaia Lied 81 Table 20 Types of surface preparation sainet esseaustiidculemesgitaandimamiotmiitleni 83 Table 21 Compressive strength of concrete in both phases oooonoccnoccnoccconcnocnnonncnnncno nono noconoconananc ran cra cra 85 Table 22 Shear test Tesults iia aiii 85 Table 23 Resultados ensayo pull Off 0 eee eecseeseseeeseecaecsecaecesecsseceeeecaeeeaeeeaeecaeecsaecaeceeeseeeseeseneeeaes 126 Bi layer diaphragm walls Experimental and numerical analysis xiv List of tables Luis Segura Castillo CHAPTER 1 INTRODUCTION 1 1 MOTIVATION An extended underground infrastructure is novvadays needed in large cities to achieve a sustainable development Bobylev 2006 The compact city strategy which aims an efficient use of space specially below ground level is one of the possible paths to allow it Durmisevic 1999 Moving activities of lesser social importance traffic or parking underground is useful in order to reduce noise pollution and polluting emissions protecting the urban environment ancie
178. nnt aw iteraci 23 2 2 2 Construction of experimental bi layer WINS ccccccccsesccceesssseessseceessesecsesseseceseceesssecsessesscsessecesssseesees 24 2 2 3 Tests and instrumentationmesiciiisi nmmitmmdietemmtmsmiitiiesisstiemmmmmtamdtusida 26 2 3 1 NUMERICALMODEL ii tenen ciment Heavens da tries 27 2 4 EXPERIMENTAL RESULTS 3 ec5cc csscescecveeee A salads SESE ARAA E Sk EAEan AKE ana a aE i anina 29 2 4 1 Material characterizatiOn csccccsesccccesssececenseessssececsensecessenaeesuececsssseseceenaecsesueecsssuesecsensessesueeecseaseseeaaes 29 2 4 2 Reliability Of inclinometers cccccccccsscceesssccceesseesesececsessssecsessesssssceceessssecessessesussecessssecsesseseeusseceessssensaes 30 2 4 3 Selection Of representative StAGCS cccscccccesssssccesscecessscccsessssscesssecsesssecsesssssceessecsesseseceessseceesseseessseeeees 31 2 5 MODELVS EXPERIMENTAL COMPARISON cccssssceceesseceseececeseececeesaececeeeeescueeeceesuececeeseeceseeeceeaaeeseeenaeens 32 2 5 1 Model adjustment camises andadas aabt ia 32 252 WGI DENGVIOUI iii eh Deki ia coa stat vei eR aa a 33 2 6 DISCUSSION DESIGN PROCESS AND COMPARISON sccccsecceceeseececeeeeeceeeeeceeueeeceesaeeceeeeeceeaeeeceeaaeeeseeaeens 34 2 6 1 Description Of COSC SEU DIOS 522223 cen dica titi AE tab ds 35 2 6 2 Results fortheoretical COS CS id di ic dic 35 AA SA O 35 2 6 2 2 B ndine MONA A A A cia ala 36 Bi layer diaphragm walls Experimental
179. nononanannnnnnnnanann anno nonnnononcnncnnnnnnnnns 97 6 3 3 Structural behaviour of the bi layer diaphragm wallS cocoooccconocnncconaccononnnocnnnnnnnonnnnccnonnnccnnnnnnnnnnnss 98 6 3 4 Overall flexural design model structural and Sectional level ooocononococcconannocnonaccnonanonononanonnnoss 98 6 3 5 Influence of the different constructions processes related to this type Of vvalls 98 6 3 6 Efficiency of the bi layer walls compared with equivalent conventional diaphragm wall alternatives 99 6 3 7 Dissemination of the results scccicctageccda cade e es ira a casat a a dd iii iia bad an ipa Aaaa Datena 99 0 4 FUTUREPERSPECTIVES s 2iinisiseiioa iniii i ia ei ad ai a asi an s 99 6 4 1 CRUCIAL TOSCO MINOS A A in Re nm ata dela el 100 642 Other researchlinesscneiueamnemsantemimatmtesmatemimeamttaimimtmlcsmetttadul 100 REFERENCE A cnsescsssccssncdtectonsscevcuteveccevsesedsewescesss donis covencsectosvvaen svseescteeewcessewedsenesdeess cones donesscnvecentaees 101 APPENDIX 1 EARLY AGES CONCRETE TO CONCRETE BOND STRENGTH ASSESSED THROUGH SHEAR AND PULL OFF TESTS A1 1 A1 2 A1 3 A1 4 A1 5 A1 6 A1 7 111 INTRODUCCI Not ts o a eta Dec O ent 112 A1 1 1 Impermeabilizaci n en pantallas CONtiNUAS uo cccccsccccsesssceceescecessececsesesecsessecscssecsesssseesssseseeeaeees 112 A1 1 2 Mareo general del provectoz iimmiuis ti dosis ani aiiin ia a a a a tun A
180. ns First according to the EHE 08 CPH 2008 formulas Ecm 8500 E fem 1 Eom Com fey Eom 9 Where Em and fem are the modulus of elasticity and the mean compressive strength at 28 days respectively and E and f are the modulus of elasticity and the mean compressive strength at time j respectively Considering that for the same characteristic strength the sprayed concrete seems to have a lower elastic modulus than the cast concrete Galobardes 2013 Malmgren 2007 an estimation of the elastic modulus for the second layer concrete was also done by equations 3 and 4 Galobardes 2013 which adjust the coefficients of the EHE 08 formulas for sprayed concrete Em 1480 3 fom 3 Bi layer diaphragm walls Experimental and numerical analysis 30 Chapter 2 Een fom El Es 4 It can be seen that the differences in the strength and the elastic modulus of the concretes from both layers are relatively small particularly for at least two ages that are shown when the wall is placed under load For example the second layer elastic modulus at 35 days is 5 above according to Eq 1 and 2 or 7 below according to Eq 3 and 4 the first layer modulus This validates the hypothesis of similar modules of elasticity in both layers 2 4 2 Reliability of inclinometers The inclinometer readings presented some systematic and individual measurement errors Data pre processing will be described using the results for Wa
181. nt buildings and parks R nk Ritola amp Rauhala 1998 At the same time placing some functions e g leisure and recreational activities traffic facilities tunnels and car parking technical maintenance facilities sewage treatment or power plants underground free aboveground space that can be addressed to recreation and social activities Durmisevic 1999 There are mainly two methods of building underground structures a excavation in an open cut from ground level and b mining or boring in tunnels These methods must be implemented without affecting existing buildings and infrastructures adjacent to the construction site and minimizing the alterations to the daily activities of the city In this city scenario where ground deformations must usually be controlled to avoid damage to existing buildings the diaphragm wall technique represents a particularly viable solution when the first of the abovementioned options is used Rodriguez Lifian 1995 2 Chapter 1 Diaphragm walls also called slurry walls EN 1538 2010 started being constructed in Italy in the 1950s Ou 2006 Puller 1994 A synthetized definition of the technique was given by El Hussieny El Hussieny 1992 describing the diaphragm walls as artificial membrane of finite thickness and depth constructed in the ground by means of a process of trenching with the aid of a fluid support This is with a dragging tools a narrow trench is excavated While the excav
182. nt de la Fuente Albert Accepted Josa Alejandro LF 2 016 Q1 CE Paper 3 Engineering Bi layer diaphragm walls Parametric study of Segura Castillo Luis Potted Structures construction processes Josa Alejandro 2 review Aguado Antonio LF 1 713 Q1 CE Paper 4 Construction and Bi layer diaphragm walls Evolution of Segura Castillo Luis Building Materials concrete to concrete bond strength at early ages Aguado Antonio 31 pp 29 37 L F 2 293 Q1 CB Conference XXXV Jornadas Bi layer diaphragm walls Early ages concrete Segura Castillo Luis Paper 1 Sudamericanas de to concrete bond strength assessed through Aguado Antonio Ingenieria shear and pull off tests Estructural Published CE ENGINEERING CIVIL CB CONSTRUCTION amp BUILDING TECHNOLOGY The research work was originally structured into four main areas a Structural level analysis b Sectional level analysis c General design and optimization and d Bonding between layers During the developing of the thesis progress was made in parallel in the four areas achieving different degree of results in each one of them As soon as enough rigorous and coherent results were obtained a paper was written and submitted for evaluation Therefore the papers interconnect the different aspects studied as it is summarized in Table 4 It can be seen that on the one hand the sectional level is less developed than the rest of the levels and on the other hand that the
183. nt to milling one day before the second phase of concrete spraying the wall was washed with a water jet Fig 42c which removes dust and loose particles produced in the milling process Besides this process also saturates the pores of the base concrete but if performed long enough in advance the surface has time to dry out leaving a dry surface but with saturated pores This final preparatory work performed on the MP walls milled surface is considered the best surface moisture condition though controversy persists over this point and contradictory results have emerged Julio et al 2004 The final step in the preparation of the EP walls epoxy bonded surface prior to concrete spraying was to place the bonding agent Fig 42d on the wall Multitek Adhesive SDH a two component water based epoxy adhesive for bonding concretes was applied following the manufacturer s instructions Likewise it is well known that moistening the surface before spraying is a technique that reduces the resistance of the bond nevertheless pre wetting the surface before applying the new concrete layer is common practice Talbot et al 1994 At the time of spraying a localized water leakage occurred at the head of the SP walls Saturated Surface therefore the surface of the diaphragm wall in this area was at all times totally saturated with water Fig 42e As water leakage was present and with the intention of confirming that this is a harmful sit
184. ntal campaign Before construction began a geotechnical study analysed the characteristics of the soil Inclinometer tubes were placed inside the walls to analyze the structural behaviour of the composite element and test specimens with poured concrete were used for material characterization as described below The bond between layers transversal displacements and anchorage loads were also measured and have been reported previously elsewhere Segura Castillo amp Aguado 2011 Segura Castillo amp Aguado de Cea 2012a 2012b Fig 6a shows the layout of the building site Standard construction methods were used to build the diaphragm walls that enclose the building site around its perimeter The figure also shows the location of the two experimental walls both running parallel to the street Within the walls the two instrumented panels are labelled Wall W35 and Wall W45 The number indicates the width of the first layer of cast concrete e g 35cm Cross sections views of these panels are shown in Fig 6b including the finished frameworks up to street level level 0 00 m the temporary anchors and the phreatic level The cross section detail of a finished bi layer wall is schematically represented in Fig 6c 0 00 m M Bond plane Vertical reinforcement 3 05 m Instrumented Panels 6 30 m 1St layer Conventional RC 21d layer Sprayed SFRC 8 30 m 10 00 m
185. ntando gradualmente y de la forma m s homog nea y continua posible de acuerdo a un ritmo de carga previamente estipulado La falla ocurre en el plano m s d bil del sistema compuesto por el disco de carga el adhesivo ambas capas de hormigones y cada una de las interfaces entre los componentes anteriores En la Fig 60 se puede ver una foto y esquema del mismo Una limitaci n de este tipo de ensayo es su relativamente poca precisi n evidenciada por las grandes variaciones de resultados que se obtienen con diferentes dispositivos Bonaldo et al 2005 Adem s los resultados dependen de algunos factores como la profundidad del testigo dentro de la capa base el espesor de la segunda capa y la excentricidad de la carga Simon Austin et al 1995 En este sentido si no se garantiza la ortogonalidad de la perforaci n la excentricidad de la carga aumentar con la profundidad de perforaci n Tambi n se cree que aumentando la profundidad de excavaci n aumenta el da o al testigo generado por las vibraciones de la broca de corte Adem s de los aspectos se alados Simon Austin et al 1995 describen otras cuestiones relativas al ensayo como propiedades de los materiales ensayados condiciones de superficie geometr a carga y efectos de disparidad de materiales el i Fig 60 Ensayo Pull off a Esquema del ensayo b Configuraci n del ensayo A1 4 RESULTADOS Y ANALISIS Los resultados se presentan en los tres apartados
186. nteriormente realizados se puede afirmar que claramente durante el disefio de la campafia experimental se subestim la influencia de los aspectos Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 123 negativos que afectan al ensayo pull off para la utilizaci n en este caso Como caso extremo de estas desalineaciones en la Fig 67 se observa la imagen de uno de los testigos extra dos mediante el ensayo pull off Fig 67 Ejemplo de desalineaciones sufridas durante la realizaci n del ensayo pull off A1 4 3 Relaci n corte pull off Se ha indicado anteriormente que varios resultados de cada serie debieron ser descartados En este apartado se analizan los resultados para los cuales se obtuvo un ensayo v lido para ambos tipos de ensayos Las parejas v lidas se grafican en las Fig 68 y Fig 69 para la preparaci n superficial por fresado y con Epoxi respectivamente Las gr ficas relacionan la resistencia a tracci n alcanzada por el ensayo pull off y la resistencia obtenida mediante el ensayo de corte y en cada una se discrimina a su vez la edad en la que se ha realizado el ensayo Soo uw Oo N o xs O T 4 4 e e N Ww 4 t Li f 1 i Resistencia a tracci n o MPa o 2 Lb gt A PF T1 O PF_T2 O PF T3 0 0 5 1 1 5 2 Resistencia a corte t MPa o Fig 68 Resistencia a tracci n vs Resistencia a corte para preparaci n por Fresado
187. nvelope of moments and a second layer amount of steel fibres so that the compound section resists the corresponding dashed line envelope of moments Although this ideal situation is difficult to reach as the steel fibres with a lower tensile strength than the traditional reinforcement strengthen the second layer it is still possible to give the second layer considerable load bearing capacity in the ultimate capacity of the compound section Bi layer diaphragm walls Experimental and numerical analysis 38 Chapter 2 0 0 i Bi layer 1 0 work 2 0 4 3 0 4 0 5 0 Depth m 6 0 7 0 8 0 9 0 10 0 I 60 40 20 0 20 40 60 40 20 0 20 40 80 60 40 20 o 20 40 Bending moment RN m Fig 19 Bending moment envelopes for the three comparative cases 2 6 3 Comparison The main characteristics and the principal results of the theoretical cases are summarized in Table 11 The following information is given for each case area of the bending moments envelope from Fig 19 discriminated by type of cross section working and total of both areas maximum horizontal displacements on the top of wall and in the central region theoretical thickness of the wall and finally whether it is a watertight design Table 11 Main results of theoretical comparison performed with PLAXIS model T W35 T BLW45 T W45 Bending moments Simple section 266 3 220 1 393 5 envelope area Compound sec
188. of the title of all papers which describe the element under study i e Bi layer diaphragm wall was removed for the sake of clarity The conference paper Conference Paper 1 is included as an appendix which complements the forth paper Besides the abovementioned journal and conference papers two chapters complete the thesis an introductory chapter and a final conclusions chapter In the first one the motivations to study the topic the objectives the general methodology and the general thesis scheme are presented In the final chapter firstly conclusions are established from the jointly consideration of the results of the different chapters and then a Bi layer diaphragm walls Experimental and numerical analysis 16 Chapter 1 brief summary of the main conclusions arisen in every chapter is presented Also further work lines both theoretical and experimental are outlined indicating future steps still needed in order to incorporate the proposed solution as a regular option Table 3 List of papers and conference papers related to the thesis Paper Journal Congress Title Authors Paper 1 Engineering Bi layer diaphragm walls Experimental and Segura Castillo Luis Ppa eee 56 pp numerical structural analysis Aguado Antonio Josa Alejandro LF 1 713 Q1 CE Paper 2 Journal of Civil Bi layer diaphragm walls Structural and Segura Castillo Luis Engineering and sectional analysis Aguado Antonio Manageme
189. of the excavation surface preparation and roughening took place to improve the bond Wall W45 was prepared by milling and Wall W35 by milling plus the addition of an epoxy bond product before spraying The second concrete layer was sprayed with a wet mix process thereby completing the structural element Part of the spraying process can be seen in Fig 8 There is general agreement in the literature that a bond Luis Segura Castillo Experimental and numerical structural analysis 25 material with a modulus of elasticity that is similar to the modulus of the adjacent concrete is desirable in the application of concrete repairs to ensure reliable performance Saucier Bastien Pigeon amp Fafard 1991 Wall amp Shrive 1988 The concrete manufacturers were therefore asked to prepare dosages with the same characteristic strength at 28 days 30 kg m of DRAMIX RC 65 35 BN steel fibres were mixed into the concrete to be sprayed in the second layer Table 5 Sequential stages of wall construction ee Day Reference name Description of the stage Scheme W35 W45 W35 W45 Casting the panel 0 0 Cast Cast Fig 7a Excavation before anchorage activation 40 33 Exc180 Exc225 Fig 7b Anchorage activation 40 33 Anc150 Anc200 Fig 7c W45 Excavation progress 0 50 ExcInt 0 W35 Excavation 3 80m 57 Exc380 Fig 7d W35 Aux anchorage 3 50m installation 57 Anc350 Fig 7e Excavation progress 1 68 61 ExcInt 1 E
190. omplete description of ground anchors for retaining structures can be found in Fang 1995 In this thesis only propped walls both in one or multiple levels are used conveniently alternating the props between active anchors struts or the structure slabs as required 1 3 1 1 Calculation methods Earth retaining structures are designed to withhold both the soil pressures and the external loads that may be applied to it being able to retransmit them to the foundation soil under controlled deformations and avoiding the collapse both of the structure and the surrounding soil Accordingly Eurocode 7 EN 2004a specifies two kind of checks to design retaining structures serviceability limit state and ultimate limit state For the first of them the code is focused in the control of the displacements of the walls and the ground adjacent to them mainly in order to avoid damage to existing adjacent buildings Regarding the ultimate limit state the code states The design of retaining structures shall be checked at the ultimate limit state for the design situations appropriate to that state For embedded walls there are mainly four groups of ultimate states modes that shall be checked a Overall stability b Rotational failure c Vertical failure and d Structural failure They can be seen schematically represented in Fig 4 Although the first three groups are checked in the cases used in this thesis they are not presented as the
191. on an overhead position If it is properly dosed and applied the sprayed concrete is a structurally sound and durable material that generally shows a good bond to the usual base materials e g concrete rock or steel The basic constituent materials of the sprayed concrete are the same as in the conventional one namely cement aggregates and water Fibres admixtures and additions are also usually incorporated into the concrete mix for spraying Despite the physical properties of a correctly applied sprayed concrete are similar to those of a cast concrete with the same composition Galobardes 2013 some differences are registered which have given place to recent studies focused on analysing and modelling these differences e g Galobardes 2013 Goodier 2000 The concrete may be sprayed by two main systems dry mix and wet mix processes being the moment when the water is introduced to the concrete mix the main difference between them Although the system started with the dry mix since the 1990s a change towards the wet mix has been registered being completely dominant nowadays in countries like Norway which has a strong tunnelling activity Franz n 1992 The reason of this change is based principally in two reasons better performance and environmental advantages A comparison of both systems can be seen in Galobardes 2013 One of the central advantages of spraying concrete is that two of the stages of the laying of the concrete pou
192. on of the maximum positive moments in the bi layer wall type also implies a small reduction in the given reinforcements As the BLss io wall type has a thinner cross section it requires on the one hand a smaller minimum reinforcement but in the other hand stronger local reinforcements to carry the design moments even though these are smaller than in the BL5o 10 alternative As an example of design the ultimate moment resistance for the BL 10 wall type is shown in Fig 26 The previously introduced design moments M are also shown in the same plot It can be seen that inequalities of Eqs 5 and 6 are satisfied in every instance It can be seen that at depths of between 0 0m and 5 5m the wall contains the SFRC layer although it is not strictly necessary as the resistance of the SS cross section itself is sufficient to cover the design moments It Bi layer diaphragm walls Experimental and numerical analysis 52 Chapter 3 would be possible to optimize the use of the fibres placing them only in the stretches where they are needed for the ultimate resistance of the cross section This is to place the SFRC layer at depths of between 5 5m and 12 5m and to place just sprayed concrete with the waterproofing additive at depths of between 0 0m and 5 5m for waterproofing purposes and to even the surface box oG eb LL KH KO 31 E 10 a El o lA A 13 14 15 16 a el e
193. on the walls on both layers as well as strain gauges and load cells at the anchorage points transverse displacements were measured by means of invar tape in order to analyze the structural behaviour of the composite element Likewise for the purpose of this study casts filled with the concrete were used for the characterization of the material and cores were extracted from the wall in order to study the bond between the two layers as described below Fig 41a shows the layout of the building site Standard construction methods were used to build the diaphragm walls that run around the entire perimeter of the building site As may be seen the building plant is not very large and is located in an urban environment with height restrictions laws The figure also shows the location used for the analysis of bonding conditions referred to as follows MP milled surface EP epoxy bonded surface and SP saturated surface Fig 41b shows a side view of the walls including the finished frameworks up to street level Five stories are planned to be built above them reaching a height of 16 5m 0 00 m 3 05 m Experimental Bi Layer vvalls 6 30 m 8 30 m 20 72 m 10 00 m Fig 41 Details of diaphragm walls a site plan b side view Luis Segura Castillo Evolution of concrete to concrete bond strength at early ages 8l 5 2 1 Preparation of specimens The first phase of the bi layer
194. on the final design comparing the use of both structural materials and materials needed to complete different waterproofing systems Chapter 4 Paper 3 Parametric study of construction processes The overall model described in the previous chapter is used to analyze and quantify the influence of different construction process in the efficiency of the bi layer diaphragm wall technique M9 Thirty numeric simulations are used to study the parameters that are grouped in two categories a specific bi layer diaphragm walls characteristics i e number of stages of spraying depth of spraying and b general diaphragm walls and construction characteristics i e wall thickness construction sequence final structure geometry Chapter 5 Paper 4 Evolution of concrete to concrete bond strength at early ages The part of the experimental campaign corresponding to the bond between layers is reported in this chapter M1 To perform it it was necessary to adapt a shear test to cores extracted from the walls M2 With the adapted test the evolution of bond strength between the two concretes at early ages was studied M3 Three preparation techniques were used milled surfaces milled and epoxy bonded surfaces and saturated milled surfaces Appendix 1 Conference Paper 1 Early ages concrete to concrete bond strength assessed through shear and pull off tests Besides the shear test pull off test were also performed to evaluate the bo
195. ond layer is applied to the sections where the moments and therefore the deformations have already developed as pointed out for the case in Fig 32b In other words large deflections will have already taken place in the area that has just been excavated and that is therefore as yet unsprayed and with no compound cross sections to assist with the displacement reduction 4 3 2 Influence of the depth of the sprayed concrete layer The depth of the excavation level when the spraying takes place and therefore the depth of the sprayed concrete layer will also influence the contribution of the SFRC layer as discussed below Fig 34 shows the bending moment envelopes for the cases with a final SFRC layer sprayed at the following depths A 12 5 m BL 60 10 BUs 4u 2S A B 11 5 m BL 60 10 BUs 4u 2S B and C 10 5 m BL 60 10 BUs 4u 2S C Once again similarities emerge between the maximum envelopes for the three cases Moreover the M gt envelopes between depth 0 0 m and 7 5 m are identical as they are defined at the stages prior to the first spraying which are the same in all three cases The area where the main differences between the three cases are concentrated is detailed in Fig 34b which also includes the position of the maximum local M X gt moment before the second section was sprayed Bi layer diaphragm walls Experimental and numerical analysis 68 Chapter 4 Firstly the bending moments increas
196. or the T BLVV45 design example The bending moment envelopes for all three cases are compared in Fig 19 Clearly the more flexible the wall the smaller its bending moments which is due to soil behaviour When the soil structure adapts to the structure this leads to changes in soil pressure tending towards values of smaller magnitude that reflect active pressure which explains the larger differences between Wall T W45 and the other two cases There are also differences between the two cast section walls with a width of 35 cm These arise after the spraying stage due to the increased stiffness of Wall T BLW34 The maximum bending moments in these two cases are 48 3 RN m for Wall T W 35 and 50 0 EN m for Wall T BLVV35 2 6 2 3 Optimum design Two zones may be differentiated in the case of Wall T BLW3S5 in Fig 19 the bending moment envelope of the stages where only the simple cross section of the cast layer is operational continuous line and the envelope of stages after spraying of the second layer where the compound cross section is working dashed line patterned fill Obviously the first layer envelope applies to both single layer walls Maximum performance could be reached if it were possible to design the reinforcements of both layers in order to resist those bending moments that they are precisely intended to withstand In other words a first layer section amount of reinforcement bars capable of resisting the continuous line e
197. ould be avoided in future It is also worth mentioning that the solution would be more effective if used on large construction sites where the switch between excavation temporary support and spraying tasks is not a significant problem A more detailed cost study and sustainability analysis should be performed to precisely quantify and compare the advantages of the solution In the authors opinion one drawback for the implementation of the technique is that it combines different relatively new structural technologies sprayed concrete SFRC waterproof concrete and methods of design diaphragm wall design through FEM Each one of them is somehow difficult to introduce as a standard technique It is natural to think that when the different techniques are combined the difficulties increase accordingly However it can also be a competitive advantage for those companies having the know how of this solution theoretical analysis and construction 6 3 SPECIFIC CONCLUSIONS Seven objectives were established in the introductory chapter of this thesis to address the main aim of the project Specific conclusions related to those objectives are established in this section For each objective the key contributions are highlighted The dots are mainly textually extracted from the partial conclusions of the thesis main body chapters The specific conclusions summarized in this section provide a general overview of the contributions in the different subject
198. oyectado y el hormig n de pantalla R A Rotura por la interfase entre el adhesivo y el hormig n proyectado Comentarios Rechazado debido a rotura defectuosa Luis Segura Castillo
199. presenting 5 of the total envelope area This smaller area is because the SFRC is sprayed when the excavation is completed depth 12 5 m so all the moments provoked by the excavation have already developed Therefore the increase in bending moments after the compound cross section is completed is exclusively due to redistribution of the internal forces that took place when the temporal supports were removed The redistribution of forces is minor as the final structural configuration 5 slabs is stiffer than the temporal supports 2 struts In cases where the spraying was in 2 stages BL 60 10 BUs 4u 2S A see Fig 32b it can be seen that the area of bi layer work was greater than in the previous case 8 In this case the increase in bending moments after the compound cross section is completed also occurred in the excavation process in addition to the aforementioned redistribution of forces Nonetheless note that the increase in bending moments in the sprayed section at depths of between 0 0 m and 7 0 m caused by the subsequent excavation is small even more so if compared with the change in moments that took place at lower depths between 7 0 m and 20 0m For example the maximum moments variation is 97 8 kN m in the upper section depth 1 5 m and 443 9 kN m in the lower section depth 10 5 m In other words most of the moments will have already developed in the area onto which the second layer will be sprayed It can be s
200. rences 105 Hoseini M Bindiganavile V amp Banthia N 2009 The effect of mechanical stress on permeability of concrete A review Cement and Concrete Composites 31 4 213 220 doi 10 1016 j cemconcomp 2009 02 003 Hsiung B C B 2009 A case study on the behaviour of a deep excavation in sand Computers and Geotechnics 36 4 665 675 doi 10 1016 j compgeo 2008 10 003 ICE 2007 Specification for Piling and Embedded Retaining Walls 2nd edition p 242 Thomas Telford Ltd doi 10 1680 sfpaervv 33580 ITA VV G 1993 Shotcrete for rock support a summary report on the state of the art in 15 countries Tunnelling and Underground Space Technology 8 4 441 470 doi 10 1016 0886 7798 93 90006 H J lio E Branco F amp Silva V D 2004 Concrete to concrete bond strength Influence of the roughness of the substrate surface Construction and Building Materials 18 9 675 681 doi 10 1016 j conbuildmat 2004 04 023 J lio E Branco F amp Silva V D 2005 Concrete to concrete bond strength influence of an epoxy based bonding agent on a roughened substrate surface Magazine of Concrete Research 57 8 463 468 Retrieved from http www icevirtuallibrary com content article 10 1680 macr 2005 57 8 463 crawler true Julio E Branco F Silva V D amp Lourenco J 2006 Influence of added concrete compressive strength on adhesion to an existing concrete substrate Building and Environment 41 12
201. research and use in engineering applications The technique has been addressed in different books for example Bentur amp Mindess 2007 Newman amp Choo 2003 There are also several manuals in this topic such as Aguado Blanco de la FUENTE amp Pujadas 2012 Gallovich Sarzalejo Rossi Perri Winterberg amp Perri Aristeguieta 2005 It is also worth Luis Segura Castillo Introduction 11 mentioning as reference the ACI state of the art report ACI Committee 544 2002 which also includes an exhaustive list of standards books and other references Reinforcement is required in cementitious materials which are brittle materials with low tensile strengths The traditional reinforcement used has been reinforcing bars appropriately located to withstand tensile stresses When fibres are used as they are discontinuous and usually randomly distributed they are not as efficient in withstanding the tensile stresses On the other hand as they are more closely spaced they have a better performance at controlling cracking This implies improvements in several properties of the SFRC like toughness ability to absorb energy after cracking impact resistance and flexural fatigue endurance ACI Committee 544 2002 Following the aforementioned differences it can be named some nowadays extensively used applications where the use of fibres has advantages over conventional bars Bentur amp Mindess 2007 classify the applications in
202. rical analysis 86 Chapter 5 throughout the neoprene sheeting These sheets are regularly changed as they are damaged in each test Moreover during the tests a very slight rotation of the sample in the clamp could be observed It is thought that the initial non linear section is because of this slight rotation and the effect of the damaged neoprene 16 P A N 00 Load KN 0 0 00 0 10 0 20 0 30 0 40 Displacement mm Fig 45 Typical shear test strength displacement graph The failure plane tended to appear on the bond plane between the two concretes due to the fact that this is weaker than any other plane within each of the concretes There are nevertheless two mechanisms which can lead to failure on another plane Since there is a bending component between the separation of the supports normal stresses appear on the bond plane under compression in the upper area and under tension in the lower area Fig 46a Normal compression on the bond plane increases the bonding value due to friction On the other hand a compressive strut is created on the plane between the load application points where perpendicular tensile stresses analogous to those in a splitting test may occur Fig 46b Therefore failure will occur on this plane if the aforementioned stresses exceed the tensile strength of the concrete before the tangential strength is overcome on the bond plane P 2 P 2 Me a a b
203. ring and Management Accepted for publication 42 Chapter 3 3 1 INTRODUCTION A widespread problem associated with diaphragm wall construction is the occurrence of leakage whenever erected in water bearing ground There are no techniques to make diaphragm walls fully watertight so a variety of alternatives all of which with different drawbacks have been developed to cope with the leakage problem Puller 1994 The waterproof system in these solutions is added to the wall after their construction is complete so it is not an integral part of the structure of the walls Considering the aforementioned points one conceivable solution would be a waterproof layer that also assumes a structural function The bi layer diaphragm wall a new slurry wall type designed to cope with the problem of watertightness in these types of walls has previously been presented by Segura Castillo et al 2013 Segura Castillo Aguado amp Josa 2013 These walls consist of two bonded concrete layers poured and then sprayed in separate stages The first is a conventional Reinforced Concrete RC diaphragm wall Once this wall attains the necessary strength soil within the perimeter is excavated and removed and the second layer this time of Sprayed Steel Fibre Reinforced Concrete SFRC and a waterproof additive is applied This paper is part of an experimental and theoretical study on bi layer diaphragm walls which has been structured into four main areas a
204. ring and compacting are merged It is particularly convenient in cases where formwork is difficult to place in areas of difficult access and where thin or with variable thickness layers and extended surfaces are needed There are several applications where sprayed concrete is commonly used Rock support is one of the main applications where sprayed concrete is used nowadays being the technique mainly developed for its use in tunnelling The advances in the technique allowed a change of role of the sprayed concrete going from being used as a provisional lining in the early days to a current use as a definitive structural lining Other relevant uses of this technique include slope stabilization structural repairs or reinforcements and metallic structures protection Regarding the last developments in the sprayed concrete technique a good insight can be seen in S Austin 2002 which presents the proceedings of the ACI SCA International Conference on Sprayed Concrete Shotcrete The conference that took place in Edinburg in 1996 was the first conference that both organizations held jointly According to Austin the research in that period was focused on specifications test methods admixtures fibre reinforcement materials the spraying process and performance Moreover after the conference it was clearly noticed the need of efforts towards the definition of test methods and the research and development of design methods to assist engineer
205. rizar y preparar la superficie y por otro lado mejorar la textura cara a favorecer la adherencia de la capa de hormig n proyectado El fresado se realiz con una fresadora de hormig n colocada en la punta de una retroexcavadora 0 00 m 3 05 m 6 30 m 8 30 m 10 00 m Fig 57 Detalle de las pantallas a Plano del sitio de obra b corte lateral Con posterioridad al fresado el d a anterior a la colocaci n del hormig n proyectado de segunda etapa se realiz una limpieza con chorro de agua Mediante esta limpieza se elimina el polvo y part culas sueltas que se producen en el proceso de fresado Adem s este proceso satura los poros del hormig n base pero al ser realizado con la suficiente antelaci n permite el secado superficial obteniendo una superficie con poros saturados pero seca superficialmente Esta es considerada la mejor situaci n con respecto a la humedad superficial si bien a n hay controversia y resultados contradictorios al respecto J lio et al 2004 En las pantallas denominadas PF Superficie Fresada esta fue la ltima tarea de preparaci n de superficie que se realiz En las pantallas denominadas PE Superficie con Epoxi se coloc antes de que se realizara el proyectado el puente de adherencia Se utiliz el adhesivo epoxi bi componente de base acuosa para uni n de hormigones Multitek Adhesivo SDH dispuesto de acuerdo a las indicaciones del fabricante
206. rmig n mediante testigos extra dos de las pantallas antes mencionadas El modelo a nivel estructural basado en elementos finitos se contrast con los resultados experimentales obtenidos El modelo seccional se tom de la bibliograf a estudiada Con el modelo estructura secci n completo se analizaron las pantallas BL El estudio muestra que la principal resistencia flexional es aportada por la primera capa el muro pantalla convencional siendo secundario el aporte de la capa de SFRC Para el rango de elementos considerados en esta tesis 35 cm a 60 cm de espesor de primera capa y 10 cm de segunda el incremento de la resistencia ltima a flexi n cuando se considera el aporte de la capa de SFRC est entre 8 y 15 Este incremento permite una reducci n en el acero de refuerzo de la primera capa de hasta un 7 0 del total del acero de flexi n y hasta cierto punto tambi n colabora con una reducci n en los desplazamientos alcanzando reducciones de hasta un 7 3 del desplazamiento m ximo Se observ tambi n que la secuencia de proyecci n es un factor clave a la hora de aprovechar la colaboraci n extra aportada por la capa de SFRC Indicaciones espec ficas se describen a este respecto Se obtuvo una buena resistencia de adherencia entre hormigones para los testigos extra dos La resistencia media medida a cada edad 2 6 y 35 d as estuvo siempre para los distintos casos por encima de 1 0 MPa M s all de los ensayos puntuales
207. rmined that adjusts the values that are obtained with failure angles other than zero In this way validated test results could be used to evaluate extracted cores with irregular bond planes 1 80 Tendency 2 line 6 days 140 O 2 days Shear strength 1 MPa 06 days A35 days T 0 30 0 20 0 10 0 00 0 10 Shear angle am rad Fig 53 Influence of failure angle on shear strength 5 3 2 7 Relationship between concrete shear strength and compressive strength Fig 54 shows the compressive strength of both concretes first and second phase depending on the shear strength in the MP series First phase concrete maintains almost constant values for compressive strength but in accordance with the results obtained by J lio et al 2006 there is an increase in bond strength as the second phase concrete gains strength According to Eurocode 2 EN 2004b the shear strength of bonds between concretes cast at different ages with no reinforcement is given by the following formula Bi layer diaphragm walls Experimental and numerical analysis 92 Chapter 5 Vrai Cfera HO 11 where c and u are factors which depend on the roughness of the substrate surface fq is the tensile strength of the concrete of lower strength and o is the stress caused by the lower normal strength through the surface that can act simultaneously with shear stress As can be seen only the concrete with the lowest strength
208. rom the general trend indicate a systematic error in the measured stage Both systematic and individual measurement error will be omitted from any future analysis in order to strengthen confidence in the measurements for this analysis Accordingly the natural reference stage i e the cast stage is omitted so the selected reference stage is specified in each analysis 2 4 3 Selection of representative stages Fig 11 shows the incremental displacements of different readings that correspond to the same structural stage in this case the elimination of the anchor of Wall W35 at a depth of 1 5 m The instant readings stages Elim150 numbers 2 amp 3 were taken immediately after releasing each of the anchors of the instrumented panel and the adjacent panels New readings were taken stages Control number 1 amp 2 at one day and at three days after release The soil shows time dependent behaviour As this study focuses on the structural behaviour of the diaphragm wall rather than soil behaviour the most representative reading of each structural stage is taken to perform the structural analysis As a rule the final reading is selected from each stage to record the largest possible deformations caused by behaviour over time 0 5 Differed 3 days PreElim150 Elim150 2 Elim150 3 0 4 Differed 1 days o w instantaneous Control 1 o N Control 2 Incr disp li
209. round space Cities 16 4 233 245 doi 10 1016 S0264 275 1 99 00022 0 El Hussieny O M 1992 A study of different aspects of diaphragm walls Tunnelling and Underground Space Technology 7 1 55 58 doi 10 1016 0886 7798 92 90113 V Bi layer diaphragm walls Experimental and numerical analysis 104 References El Razek M E A 1999 New method for construction of diaphragm walls Journal of Construction Engineering and Management 125 4 233 241 Retrieved from http ascelibrary org doi pdf 10 1061 ASCE 0733 9364 1999 125 3A4 233 EN 2004a Eurocode 7 Geotechnical design Part 1 General rules European Committee for Standardization EN 2004b Eurocode 2 Design of Concrete Structures Part 1 1 General Rules and Rules for Buildings European Committee for Standardization EN 1538 2010 Execution of special geotechnical work Diaphragm walls European Committee for Standardization CEN Fang H Y 1995 Foundation Engineering Handbook p 942 Chapman amp Hall FIB 2010 F d ration Internationale du B ton Model Code 2010 first complete draft vol 1 Bulletin 55 Lausanne Switzerland Foye K C amp Jaoude G A 2004 Limit States Design of Deep Foundations on Design and Construction of Deep Foundations p 245 doi 10 5703 1288284313262 This Franz n T 1992 Shotcrete for underground support a state of the art report with focus on steel fibre reinforcement Tun
210. rproofness is also considered Luis Segura Castillo Introduction 5 The objectives of the research vvere Ol 02 03 04 OS 06 07 1 3 Corroborate the viability of the proposed solution Assess the bond strength reached between the concrete layers Assess the structural behaviour of the bi layer diaphragm walls Develop an overall flexural design model structural and sectional level Quantify the efficiency of the method when compared with equivalent conventional diaphragm wall alternatives Study the influence of the different constructions processes related to this type of walls Disseminate the results THESIS BACKGROUND As it was seen in the previous section many and diverse disciplines have to be combined in order to materialize this new structural element The more important ones are shown in Fig 3 where some of the interconnections are schematically represented I l Soil structure interaction Soil Behaviour N No THESIS BACKGROUND es EE ee I Fig 3 Main knowledge areas needed to materialize the bi layer diaphragm walls A brief introduction to these topics which conforms the background of the thesis is presented in this section with the following aims a define the terminology used in the thesis b summarize key points in every topic and c guide the reader into specialized bibliography Bi layer diaphragm walls Experimental and numerical an
211. rry wall made of two bonded concrete layers poured and then sprayed in separate stages The first is a conventional reinforced concrete RC diaphragm wall Once this wall attains the necessary strength soil within the perimeter is excavated and removed and the second layer this time of sprayed concrete with steel fibres SFRC and a waterproof admixture is applied This research work has been structured into four main areas a Structural level analysis b Sectional level analysis c Bonding between layers and d General design and optimization The main objective of this paper is to corroborate the structural level behaviour of the bi layer diaphragm walls both experimentally and Luis Segura Castillo Experimental and numerical structural analysis 23 numerically i e the first of the aforementioned areas To do so a methodology with three components was followed a demonstrate the viability of the proposed solution by reporting on the experimental campaign to assess the structural behaviour of the bi layer walls b develop a Finite Element Model FEM capable of predicting the structural behaviour of the bi layer diaphragm walls and c assess the structural contribution of the second layer with the cast RC wall through a theoretical example of use 2 2 EXPERIMENTAL PROGRAM 2 2 1 General Information The structural behaviour of various bi layer walls at a building site in Barcelona Spain was analysed in a full scale experime
212. rte LCB a Esquema del dispositivo b Configuraci n del ensayo La raz n de ser de la elecci n de este ensayo responde a dos motivos por un lado permite ensayar a corte testigos extra dos por otro lado permite ensayar m s f cilmente uniones irregulares entre capas problema reportado para el ensayo gillotina Delatte Jr et al 2000 ya que deja un peque o espacio entre los bordes de introducci n de carga Sin embargo al tener esta separaci n la componente de flexi n que act a en la uni n a ensayar es mayor La tensi n rasante que aparece en la zona de la interfase correspondiente a la junta se calcula seg n la f rmula t P 2 S 12 d nde t tensi n de corte en MPa P fuerza m xima de falla en N S rea de la secci n transversal del esp cimen en mm2 A1 3 4 Ensayo pull off El ensayo pull off es uno de los m todos de ensayo a tracci n m s com nmente utilizados para evaluar la adherencia entre dos hormigones De acuerdo a la norma ASTM 2009 el procedimiento de ensayo Bi layer diaphragm walls Experimental and numerical analysis 118 Appendix 1 consiste en pegar mediante un adhesivo un disco de carga a la superficie de la segunda capa Luego de que el adhesivo se ha endurecido un dispositivo de carga se fija al disco de carga y se alinea de forma de que la fuerza se ejerza en direcci n perpendicular a la superficie a ser evaluada La fuerza aplicada por el dispositivo se va aume
213. ructural analysis 29 Geogrid structural elements and node to node anchor elements were used to model the body and the free length respectively of the ground anchors Their properties are shown in Table 6 Two external loads were considered in the model 3 0 kN m was placed over the pavement to represent its extra weight and 50 0 kN m on the opposite side of the street to represent the building weight The street width is shown in Fig 6 2 4 EXPERIMENTAL RESULTS 2 4 1 Material characterization Table 9 shows the material characterization for both layers of the wall and provides the mean compressive strength of the concrete and the elastic modulus of each layer The concrete strengths of the first layer which were already reported in Segura Castillo amp Aguado de Cea 2012a corresponds to Wall W35 Other age strengths and Wall W45 values are also reported in Segura Castillo amp Aguado de Cea 2012a Table 9 Compressive strength of concrete in both phases Age of concrete days fem N mm2 Ecm N mm2 Ist layer 2nd layer 2nd layer Ist layer 2nd layer Ist layer 2nd layer Eq 1 amp 2 Eq 1 amp 2 Eq 3 amp 4 28 34 82 27756 87 2 38 94 30 99 28703 27033 22047 91 6 39 05 39 22 28728 29012 24826 120 35 39 97 45 40 28929 30314 26726 Values calculated according to the concrete maturity equations Neville amp Brooks 2010 The modulus of elasticity was calculated from the characteristic strength by two mea
214. s 6 3 1 Viability of the proposed solution e The construction of the experimental walls showed that the bi layer diaphragm walls are viable with the present state of the construction technology e Besides the correct structural and bonding behaviour more experiences are needed to corroborate the waterproofness and sectional response of the walls 6 3 2 Bond strength reached between the concrete layers e A test was proposed for determining shear strength between concretes poured in different layers The adopted test allows the testing of extracted cores being therefore suitable to be used on real structures It yields valid results with a coefficient of variation lower than 15 e Shear strength in the milled surface increases with the age of second layer concrete fitting to the maturation formulas e Shear strength assessed with the proposed test is independent of the direction in which the milling of the bond surface is carried out Bi layer diaphragm walls Experimental and numerical analysis 98 6 3 3 6 3 4 Chapter 6 As expected the saturation with water of the surface at the time of spraying the second phase diminishes the bonding capacity of the link The in situ preparation of the surface with epoxy shows a wider dispersion of results compared with the milled surface with a coefficient of variation of up to 57 in the worst of cases This behaviour may be attributed to the inherent difficulties in the
215. s designed to cover only the SS design moments the second layer does not provide the additional bending strength to the CS cross section to cover the moments developed after the second layer is sprayed Therefore the SS sections should be designed to cover the M i and part of the M a moments The second reason is that the minimum M gt j given by the minimum reinforcement already covers a part of the yw design moments Even though a reduction in the RC reinforcement is confirmed for both wall types 2 1 and 2 5 it appears insufficient to compensate for the extra technologies and consumption of materials to build the bi layer solutions Nonetheless the complete waterproof system becomes an interesting solution when particular conditions are taken into account such as basement space limitations or if continuous maintenance wants to be avoided Future work should include a parametric study to evaluate by means of the two level comparison presented in this study the influence of the general condition and wall design on the profitability of the bi layer wall type Luis Segura Castillo Structural and sectional analysis 55 3 8 ACKNOWLEDGEMENTS Funding was made available from the Spanish Ministry of Education and Science through Research Project BIA2010 17478 Procesos constructivos mediante hormigones reforzados con fibras Luis Segura Castillo is grateful for the Fellowship awarded by the FPU Spanish Research Program AP2010 3789
216. s in the design of elements using this technique Luis Segura Castillo Introduction 13 The increase in those efforts can be confirmed observing the amount of congresses that took place and journal papers published in the subsequent years An idea of the direction of those efforts can be seen in Celestino amp Ishida 2009 where the work done by the ITA working group on sprayed concrete use was presented It includes a report that compile information provided from different ITA National Groups where it can be highlighted the trends followed in the last years i e towards the wet mix system the progress in the use of alkali free accelerators the use of sprayed concrete for permanent linings and the substitution of wire mesh by fibres The report also compares the codes standards and guidelines adopted in different countries It remarks that different concepts were adopted in different countries for the design with sprayed concrete Therefore despite the efforts and progress made the sprayed concrete technique is still not mentioned in some of the more important concrete design codes at European level e g Model Code 2010 FIB 2010 Eurocode 2 EN 2004b which may be one of the reasons for sprayed concrete not being extendedly used as a structural material However several standards and recommendations address different aspects of this technique They can be grouped as they are mainly related to two technical committees
217. sidering the properties of both the soil and the retaining wall and its interaction However like the classical methods it is still not able to take into account the arching effect El Overall stability Fig 4 Typical ultimate states modes in embedded walls Fuente Eurocode 7 EN 2004a The empirical approach was mainly used in the English speaking countries It is based on comparing the characteristics of the project with monitored case histories results of resembling excavations Bi layer diaphragm walls Experimental and numerical analysis 8 Chapter 1 Finally from the 1970s with the advance in the numerical methods and the generalization of computer calculations the finite element method started to be used in geotechnical problems The soil is considered as a continuum which is discretized in several finite elements Being both the structure and the soil modelled it allows a more comprehensive representation which includes its interaction and the arch effect Also information of all the soil under study is obtained Sanhueza Plaza 8 Oteo 2007 1 3 1 2 FEM in embedded walls problem Potts amp Zdravkovi 1999 and Ou 2006 can be named as example of books focused on geotechnical analysis through the FEM method A positive aspect of the method is that in theory as more factors are appropriately considered in the FEM models the accuracy of the results would be higher than the previous ones On the ot
218. siguientes pudi ndose consultar la totalidad de los resultados referidos al ensayo de corte Apartado A 1 4 1 as como un an lisis de los mismos en Segura Castillo amp Aguado 2012 A1 4 1 Resultados principales del ensayo de corte En la Fig 61 se grafica el valor medio obtenido para cada edad de ambas preparaciones superficiales En la Fig 62 se grafica la dispersi n est ndar calculada para cada edad de ambas preparaciones superficiales A1 4 1 1 Preparaci n por fresado Para las distintas edades se obtuvieron dispersiones homog neas entre 0 05 MPa y 0 18 MPa Se puede observar claramente el aumento del valor de la resistencia a corte al aumentar la edad En la Fig 61 se traza a su vez la curva de mejor ajuste se presenta la ecuaci n de dicha curva y su valor de R Dicha ecuaci n es de la forma de la funci n de Plowman modificada Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 119 TS A B log madurez 13 d nde t tensi n de corte A B constantes a determinar madurez t rmino definido por la funci n de Nurse Saul madurez X T Ty At 14 d nde To temperatura datum usualmente 10 C At intervalo de tiempo Por lo que en concordancia con el estudio de Delatte et al 2000 la evoluci n de la resistencia a corte para la segunda fase realizada con hormig n proyectado se ajusta perfectamente al modelo de maduraci
219. sing the adequate spraying sequence a first layer reinforcement reduction larger than 10 Rg m can be obtained in every configuration and sequence reaching a reduction up to 21 7 kg m in the best case This represents a percentage reduction of 7 0 of the total bending reinforcements The spraying sequence is a relevant parameter in the design of the bi layer walls In cases where moments do not increase after the excavation process is completed the spraying should be performed during the excavation Otherwise spraying must be done after finishing the excavation process In every case the sooner the spraying is performed the larger the bending moments developed It is confirmed that the displacements are governed by the thickness of the first layer being practically identical for each combination of final structure configuration and construction sequence A minor reduction in displacement is registered when the second layer is included The maximum reduction obtained 0 6 mm represented a percentage reduction of 7 3 As it was said in section 6 3 3 the majority of the deflections caused by the increase in the bending moments have already taken place when the excavated sections are sprayed Efficiency of the bi layer walls compared with equivalent conventional diaphragm wall alternatives A comparison of different complete systems to deal with leakages i e conventional diaphragm wall plus an independent waterproofing method confirmed that
220. st layer thickness W1 55 55 cm Ist layer 60 60 cm Ist layer 2nd layer thickness W2 0 Mono layer type 10 10 cm 2nd layer Construction sequence CS BUs Bottom Up with struts BUa Bottom Up with anchorages TD Top Down Final number of underground levels 2u Infrastructures NU 4u Dwelling basements Number of spraying stages NS OS Mono layer type 1S 1 stage spraying 2S 2 stage spraying 4S 4 stage spraying Depth of sprayed concrete layer in the M Mono layer type last stage DS A Depth 12 5 m B Depth 11 5 m C Depth 10 5 m The two final configurations both of which can be relatively common are shown in Fig 28 The first alternative see Fig 28a is a structure with an intermediate slab apart from the base slab and the ground level slab It could for example be an underground station where the upper enclosure would be for the station service area and the lower one for the platforms tracks and other railway equipment The second alternative see Fig 28b has three intermediate slabs besides its top and bottom slabs which might for example be an underground car park at four levels Fig 28 Final construction design a 2 levels 2u b 4 levels 4u Fig 29 shows examples of the construction sequence A generic BU sequence for the 4u alternative is represented in Fig 29a The temporary supports are generically represented with a horizontal arrow in the Bi layer diaphragm walls Experimental and numerical ana
221. stage the wall has the CS cross section at depths of between 0 0m and 3 0m Therefore until this stage is complete the bending moments are resisted exclusively by the SS cross section The envelope of these moments is represented with an unbroken bold line The changes in bending moments of stage Exc 2b are shown in Fig 25c The increase in bending moments at depths of between 0 0m to 3 0m can now be withheld by the CS cross section The maximum moments that develop once the CS cross section is completed are referred to as M and its envelope is represented with a bold dashed line as shown in Fig 25d This figure represents the situation after spraying the second stretch Spray 2b in which the two kinds of envelopes may be seen 3 5 SECTIONAL RESULTS The design criteria set the ultimate moment resistance as equal or greater than the design moment of each cross section My gt Ma This particular criterion is used for the dimensioning of the main vertical reinforcement which accounts for the differences introduced by the various wall types analysed in this study Therefore secondary reinforcements e g for transversal stresses or time dependant effects are neglected in this study as they are considered the same for all three wall types The shear force and its reinforcement is also neglected as it is not usually a determinant in the design of the walls Reinforcement of the RC layer involves a a symmetric reinforcement on bo
222. steel fibres used in the SFRC layer under steel In both cases concrete and steel cast and sprayed materials were differentiated The consumption of extra materials of the DC system as it is of a different class is not considered in the table Although the dosages placing procedures and cost are not the same for the different types of materials this simplification allows a simple first approach to compare the different systems With regard to material consumption the ML wall type is the one with minor outlay in every material category Considering that maximum displacements are similar for all systems and that this one in particular has the smallest ones and finally that this system has the smaller thickness this system is undoubtedly the most appropriate whenever waterproofness is not required The bi layer wall types achieve a reduction in the RC layer steel reinforcement The percentage of reduction in this layer compared with the ML p alternative is 2 1 for the BLeo 0 wall type and 2 5 for the BLss jo wall type However the steel increment in the SFRC layer exceeds the reduction reached in the RC layer The percentage increment compared with the MLso wall type is 8 0 for the Unif system and 4 5 for the Opt bi layer system As the material and labour costs per cubic meter of sprayed SFRC are higher than the cost of RC the structural system of the bi layer solutions is not favourable in this case Compar
223. structural behaviour of the wall the structural advantages provided by the collaboration of the steel fibre reinforced concrete SFRC layer 96 Chapter 6 and the bond between layers General conclusion of the research is presented in the following section Subsequently specific conclusions in response to the different objectives are presented Finally general lines for future work needed in order to complete de development of the bi layer diaphragm walls are introduced 6 2 GENERAL CONCLUSIONS In general terms it can be said that the research performed laid the foundation for the development of the bi layer diaphragm wall technique which is a promising solution for the leakage problem of diaphragm walls The advantage of the method resides in the efficient use of the materials of a diaphragm wall that needs to be waterproof A double function structural and waterproofing is assigned to the second layer which therefore is able to collaborate with the overall structural response In this thesis a complete flexural design method based on an uncoupled structural section analysis was established chapter 3 Furthermore the structural model based on a FEM model was contrasted with the experimental walls chapter 2 Various theoretical cases where analysed through this method to quantify the efficiency of the proposed solution For the geometrical ranges of the elements considered in the thesis first layer between 55 cm and 60 cm an
224. supports of the other construction sequences Two consequences can be mentioned Firstly as with the BUa case but in a contrary sense in this case the supports are stiffer and therefore the displacements and bending moments are smaller Secondly that the local maximum moment produced by the intermediate slab depth 6 0 m reaches positive moment values There is an area of bi layer work that can be seen around the positive maximum The positive moments at this depth developed after the second layer had been sprayed Therefore the additional compound cross section strength provided by the second layer can be harnessed this time with the SFRC working in compression However note that there are high shear forces near the support together with the peak of moments Therefore the shear strength of the element and the debonding risk between layers should be evaluated with particular attention Note also that there is no internal redistribution of forces in the TD sequences as in these cases the walls are directly supported by the finished slabs Bi layer diaphragm walls Experimental and numerical analysis 72 Chapter 4 Fig 38 shovvs the horizontal displacements for the corresponding cases shovvn in Fig 37 It also includes the displacements of the respective mono layer cases ML 60 0 BUa 2u 0S M for the BUa case ML 60 0 BUs 2u 0S M for the BUs case and ML 60 0 TD 2u 0S M for the TD case 0 0 0
225. tado fue necesario extraer los testigos con una distribuci n irregular Se eligieron estas zonas para disponer la m quina de extracci n de testigos lo m s perpendicular al plano medio del paramento del muro Luis Segura Castillo Early ages concrete to concrete bond strength assessed through shear and pull off tests 117 A1 3 3 Ensayo a corte De los ensayos descriptos en la introducci n se eligi para el ensayo a corte el ensayo LCB modificado En la Fig 59 se observa un esquema del dispositivo y una fotograf a del ensayo El ensayo se basa en la norma NLT 382 08 2008 pensada para la evaluaci n de la adherencia en capas de firme compuestas por materiales bituminosos Mir Recasens et al 2005 materiales mucho m s d ctiles que el hormig n y cuyas propiedades son mucho m s sensibles a las variaciones de temperatura Por tal motivo se suprimi la c mara de control de temperatura y se redujo la velocidad de desplazamiento del pist n de carga de 2 5 mm mm a un valor en el orden de los usados para ensayos de corte en hormig n 0 25 mm min Mirsayah amp Banthia 2002 Ray et al 2005 Wall amp Shrive 1988 Para reducir la concentraci n de tensiones en los apoyos se coloc una l mina fina de neopreno entre el dispositivo y el testigo En la realizaci n de los ensayos se ha utilizado una prensa hidr ulica con control de desplazamientos Apra Apr T 1 2 94 mm 1L 2 94 mm Fig 59 Ensayo de Co
226. tage urface moisture before second stage concreting before second stage concreting concreting Bonding agent No bonding agent Epoxy adhesive No bonding agent Bold values indicate the distinctive preparation of each type of surface The cores for the study of bonding between the layers were extracted from the wall one day before the scheduled date for their test They therefore retained the same curing conditions as the rest of the element for as long as possible Since tests were planned at different ages the extractions were also carried out at different ages Core extraction was performed in the MP and EP areas when the second phase concrete was 1 day 5 days and 34 days old 5 cores from each of the areas were extracted at each age In the SP area only 4 cores were extracted at the age of 34 days Some cores mainly those extracted at the earliest ages broke along the bond plane at the moment of extraction Various agents intervene in studies that take place under real working conditions on site e g Promoter Constructor Laboratories Researchers The circumstances under which this work was carried out made it very difficult to modify the experimental programme as initially planned In addition especially at the first age t 2 days there was less than one day in which to perform the core extractions and the rest of the experimental measurements At that age it was therefore not possible to extract more specimens to replace the six
227. ted structural and functional improvements have been described in this paper Structural sections are usually designed as a conventional diaphragm wall to which a second waterproofing layer may be added afterwards However in this proposed solution the second layer is sprayed and bonded to the first one during construction so that it fulfils a structural role Hence this procedure leads to an optimized section with improved watertightness An experimental and theoretical study applied to a full scale case has been performed which examined two bi layer walls of different cross sections The experimental results of readings from inclinometers embedded in the walls were reproduced with a numerical model running on the PLAXIS FEM program This model was subsequently used to quantify the structural contribution of the second layer with the cast RC wall The following conclusions have been drawn 1 An experimental campaign involving bi layer walls was successfully conducted The results support the viability of this solution and demonstrate the monolithic behaviour of the combined layers The results however only apply to this particular case study and care must be taken directly extrapolating to other examples without further research 2 A comparison of the results from the calibrated model and from the experimental campaign demonstrated a very good correlation which validated the model It considered cross sectional changes in the stages after spr
228. th sides of the wall with the minimum mechanical reinforcement Asmin according to the EHE 08 code CPH 2008 and b one extra reinforcement per side of the Wall one for the positive moments A and another one for the negative ones A to cover the extra moment that the minimum reinforcement does not cover The addition of both areas Ar was used in the calculations for cross sections in which both reinforcements were present Only tensioned bars were used in the calculation Two ultimate moment resistances whether or not we consider the SFRC layer were obtained for the bi layer wall types one for the SS M and another for the CS cross section MF gt In this way the design condition for the bi layer walls can be differentiated according to the type of section that is active at each instant establishing that every cross section must at every instant simultaneously satisfy both relationships given by the following inequalities SS SS MS gt M 5 6 CS CS Luis Segura Castillo Structural and sectional analysis 51 The values of the reinforcements obtained for the three wall types are shown in Table 14 The following information is given for each alternative bar diameter d and bar spacing s expressed in the form d s the position of the reinforcements z and z see Fig 26 and the ultimate moment resistance of the SS and the CS cross sections T
229. the American Concrete Institute ACI and the European Federation of Producers and Applicators of Specialist Products for Structures EFNARC A summary of standards related to both groups can be found in Galobardes 2013 The works related to sprayed concrete performed for this thesis were carried out mainly following the EFNARC guidelines 1 3 5 Bond between concretes Bond strength is a key parameter in the performance of structures composed by concrete placed in different times Good bond strength is needed in order to allow the structure to behave monolithically and to effectively mobilize the strength of the different components A brief literature review about bond between concretes written with a special focus on its application on the bi layer diaphragm walls can be found in Segura Castillo amp Aguado de Cea 2012a Included in section 5 1 of this thesis 1 4 GENERAL METHODOLOGY The general work methodology is based on a combination of experimental works laboratory and field and numerical tools where different data obtained in the experimental campaigns is used as an input for the models and to validate them All along the thesis a special emphasis was placed on the SFRC layer contribution The general methodology can be broken down into a series of tasks as follows Ml An on site full scale experimental campaign where the bi layer walls were constructed was performed during the construction of a building located in Barce
230. the bi layer walls are efficient only if waterproofness is needed This is a conventional diaphragm wall alone without waterproofing is always more economical than an equivalent bi layer alternative The comparison of the material consumption concrete and reinforcement steel of chapter 3 showed that the bi layer system does not reduce the total use of materials of the complete waterproof systems reaching in the best cases similar consumption of materials Therefore the final cost is probably still higher for the bi layer system as the material and labour costs per cubic meter of sprayed SFRC are higher than the cost of RC of the conventional diaphragm walls Dissemination of the results Two papers were published one accepted for publication and one is under the second review in international journals JCR JOURNALS Q1 each of them corresponding to each of the chapters of the thesis main body With the followed strategy feedback was obtained from the reviewers in the publishing process regarding the more advanced parts of the work at the same time as the rest of the thesis work was being completed One paper was published in a South American international conference It followed the strategy of making both the research and the PhD candidate known in the region where he intends to develop his career as university professor FUTURE PERSPECTIVES Further research is necessary to complete the advances of the bi layer diaphragm w
231. tion 57 9 kKN m m Total 266 3 278 0 393 5 Horizontal displacement Top 0 63 0 64 0 99 mm Centre 4 99 4 97 4 70 Theoretical thickness cm 35 45 45 Waterproof wall NO YES NO Regarding the displacements of Wall T W35 the results are even practically comparable to the bi layer solution with differences of below 2 In return a reduction in the reinforcement needed for the simple cross section can be achieved In the case of Wall T 35 the simple section must withstand 266 3 kN m m whereas in the T BLW35 the value to be withstood is reduced to 220 1 kKN m m which implies a reduction of 17 Although it is impossible to take full advantage of this moment reduction reductions in the reinforcement of up to 10 have been obtained in preliminary calculations It should not be forgotten that the main advantage of the bi layer wall is its watertight design The single layer Wall T W45 was considered to compare two solutions with the same final thickness Besides the previously mentioned considerations regarding the addition of the second layer the slender solution is advisable for all comparative parameters listed in Table 11 except for deformation in the centre of the wall Thus the thickness of the cast layer may be reduced insofar as is permitted in accordance with Luis Segura Castillo Experimental and numerical structural analysis 39 the maximum displacement Then the advisability of using the bi layer solution may be analysed a
232. tolerances 25 cm in contact with every diaphragm wall would be lost In a vehicles parking basement the additional space would represent the difference to afford a parking place intended for a luxury car In addition to the space lost a drained cavity may in the worst case hide dangerous leakages and even structural problems Puller 1994 Bi layer diaphragm walls Experimental and numerical analysis 4 Chapter 1 In below grade structures in general it can be said that leakages are originated mainly by poor design and workmanship rather than the selection of materials Therefore the key to avoid leakages resides in the design and implementation Shohet amp Galil 2005 Beyond the waterproof method the early selection of the type of wall the construction sequence and the temporal and permanent use of the retaining structure have a positive effect on the final cost Hence client designer and contractor should all be involved in the project at an early stage Gaba Simpson Powrie amp Beadman 2003 A holistic vision of the project requirements should be adopted to achieve optimization in which the costs of material consumption the final dimensions of the wall maintenance requirements and construction complexity should all be evaluated throughout its entire life cycle Furthermore to achieve sustainable design the best strategy is to consider environmental aspects also right from the start of the design process Kurk amp Eagan 20
233. uation an on the spot decision was taken to test this zone too Bi layer diaphragm walls Experimental and numerical analysis 82 Chapter 5 Fig 42 Main steps in the production of the specimens a milling the wall built in the first phase b finished surface c water jet washing d placing the bonding agent e area of water leakage f spraying the second layer of concrete Table 20 summarizes the main processes characterizing the three types of surface preparation The second phase concrete was sprayed with a dry mix process Fig 42f thereby completing the structural element There is general agreement in the literature that a bond material with a modulus of elasticity similar to the adjacent concrete is desirable in the application and for the performance of concrete repairs Simon Austin et al 1999 Wall amp Shrive 1988 As the two phases are placed at different times different moduli of elasticity develop in both concretes as they gain strength Therefore it is understood that equal moduli should be achieved throughout the service life of the structure Various rules including Eurocode 2 EN 2004b establish a relation between the modulus of elasticity and concrete strength Therefore the firms that supplied the concrete were asked to prepare dosages with the same characteristic strength at 28 days The dosage of the sprayed concrete design Table 19 was based on proposals made by Garc a et al Garc a Vicente
234. uccesuaetcceceecevogonace Sacecccuuoestadouccrvsaonecede ddcaseacerbaseucesevedtecste 84 5 3 1 Mechanical characterization reSults cccccccssscccsssscecessececsensecesssccecsenecesenaecsseeeeecseueeesseaaessseeeeseeaeeeeeea 84 53 2 Shedrtestresults ial uti iaia A A A O TETE 85 5 3 2 1 NR A mts et mt E ei 85 5 3 2 2 Sh arstress dependine on agent id 87 5 3 2 3 Shear stress N SA TUTTO surfaces seis a ai 88 5 3 2 4 Shear stress on epoxy bonded SUrfacCes ossen n A aa E a A E A EEEak 88 5 3 2 5 Mihe direction ui ii EA r a A E AAE AR ARRE 89 5 32 01 Shear angle a a aida E oo lo dodo es 90 5 3 2 7 Relationship between concrete shear strength and compressive Strength ooooocninccnncconocnconcnonnncnononos 91 Bid CONCLUSION Sorteo E E E E E E ca ed E TE NTE 92 55 ACRNOMNLEDGMENTS tir naaitecamai ei EREE E ATE E REE miren E ARETE E EREE 93 Luis Segura Castillo Table of contents ix CHAPTER 6 CONCLUSIONS AND FUTURE PERSPECTIVES sccccscccsccccccecccecececcceeccesceescessscessceescesssceseeeeseeses 95 6 1 INTRODUCTIONS sciences Bei Gestions o eases 95 6 2 GENERAL CONCLUSIONS sis misma di ds 96 6 3 SPECIFIC CONCLUSIONS ata doten eR desar aie Annan dian eka ee aie 97 6 3 1 Viability of the proposed solution cccccceesssccceessecsssssecesssesssussseceessssesssesscsasseceesssseceessseesseeceesssseceaaes 97 6 3 2 Bond strength reached between the concrete layers ococococoocoocnno
235. urface Likewise it examines the influence of the milling direction and the compressive strength of the constituent concretes on the actual bond strength The study was conducted at a real construction site Some of the techniques that are commonly associated with the construction of diaphragm walls do not according to the literature always offer the best results However it was decided to maintain these techniques to simplify the implementation of the new structural typology Thus for example milling was used instead of sand blasting with better results according to J lio Branco amp Silva 2005 and adhesive epoxy instead of modified cement with better results according to Momayez et al 2005 Bi layer diaphragm walls Experimental and numerical analysis 60 Chapter 5 This article represents an important step forward in the development of bi layer diaphragms contributing knowledge on the bond formed at early ages by concrete that is sprayed over an existing layer especially over a milled concrete surface one of the most usual practical methods of in situ preparation It is useful for engineers that design bi layer elements and structural repairs involving sprayed concrete reinforcements in general 5 2 EXPERIMENTAL PROGRAM An experimental campaign was developed to test the proposed methods The previously described bi layer walls were constructed at a building site located in Barcelona Spain Inclinometers were placed
236. ved from http www concrete org PUBS JOURNALS AbstractDetails asp ID 1376 Terzaghi K Peck R B amp Mesri G 1996 Soil mechanics in engineering practice p 592 New York John Wiley amp Sons Luis Segura Castillo References 109 Tu L amp Kruger D 1996 Engineering properties of epoxy resins used as concrete adhesives ACI Materials Journal 93 1 26 35 Retrieved from http www concrete org PUBS JOURNALS OLJDetails asp Home MJ amp amp I D 9793 UNE EN 12390 3 2003 Ensayos de Hormig n Endurecido Parte 3 Determinaci n de la resistencia a compresi n de probetas Madrid AENOR Asociaci n Espa ola de Normalizaci n y Certificaci n Retrieved from http www aenor es aenor normas normas fichanorma asp tipo N codigo N00438088PDF Si UkGkxtKno80 UNE EN 14488 1 2006 Ensayos de hormig n proyectado Parte 1 Toma de muestras de hormig n fresco y endurecido Madrid AENOR Asociaci n Espa ola de Normalizaci n y Certificaci n Retrieved from http www aenor es aenor normas normas fichanorma asp tipo N codigo N00356828PDF Si UkGkRdKno8o0 Wall J amp Shrive N 1988 Factors affecting bond between new and old concrete ACI Materials Journal 85 2 Retrieved from http www concrete org PUBS JOURNALS OLJDetails asp Home MJ amp ID 2329 Wang J H Xu Z H amp Wang VV D 2010 Wall and Ground Movements due to Deep Excavations in Shanghai Soft Soils Journal of Geotechnical and
237. xcInt 1 End of excavation 6 30m 78 71 Exc630 Exc630 Fig 7f Spraying of second layer 85 78 Spray Spray Fig 7g Construction of base slab 91 84 Base Base Fig 7h W45 Control measurement 1 90 Control 0 Fig 7h W35 Before aux anch elimination 97 PreElim350 Fig 7h W35 Aux anchorage 3 50m Elimination 1 97 Elim350 1 Fig 7i W35 Aux anchorage 3 50m Elimination 2 97 Elim350 2 Fig 7i Construction of lower slab 100 93 Slab Slab Fig 7j Before anchorage Elimination 109 102 PreElim150 PreElim150 Fig 7j Anch Elimination 1 109 102 Elim150 1 Elim200 1 Fig 7k Anch Elimination 2 109 102 Elim150 2 Elim200 2 Fig 7k Anch Elimination 3 109 102 Elim150 3 Elim200 3 Fig 7k Control Measurement 2 110 103 Control 1 Control 1 Fig 7k Control Measurement 3 112 105 Control 2 Control 2 Fig 7k Bold letters indicate stages where inclinometer readings were performed Underlined letters indicate stages selected for the comparison W45 W35 specified only where necessary if stage differs in each wall Table 6 Anchorage properties Initial tensile Total Bulb Cross section Elastic Anchor load Depth length length area modulus angle KN m m m mm KN mm W35 500 1 5 20 0 14 0 563 92 198 46 30 W35 Auxiliary 300 3 5 13 5 8 5 281 96 198 46 30 W45 500 2 0 20 0 14 0 563 92 198 46 30 The required thickness of the second layer was 10 cm however layer thicknesses ranging from 9 cm to 17 cm were d
238. ype are detailed in Fig 24b in which light grey lines indicate the moment of the representative stages of each excavation stage The interval between the envelopes previously introduced in Fig 24a MP and MS is highlighted with slanting lines This area represents the increase in the moments after spraying the SFRC layer i e where the CS cross section is working As stated in Segura Castillo Aguado amp Josa 2013 the highlighted area represents the potential use of the bi layer wall since it is possible to cover these moments with the resistance of the CS section It can be seen that for the depths where the CS section is present a significant portion of the bending moments are developed after the SFRC layer has been sprayed These increases range from 30 to 269 at depths of between 2 5m and 11 0m with an average increase of 123 in the design moment of those depths after the SFRC layer is sprayed The value of the area within the SS cross section envelope is represented with a solid bar graph to compare the three wall types in Fig 24c The value of the area of the CS cross section as shown in Fig 24b is also plotted slanting lines It may be noted that the potential of use of the SFRC layer covers approximately 25 of the area of moments The bending moments plots of the BL o 10 wall type at depths of between 0 0 m and 5 0 m for the stages from Exc 2a to Spray 2b are shown in Fig 25 For each plot dark li

DOCTORAL THE SIS - Tesis doctorales en red

Contents

Download Pdf Manuals

Related Search

Related Contents