Citation
Experimental investigation of ultra-high performance concrete for precast decked bulb-tee bridge girder connections

Material Information

Title:
Experimental investigation of ultra-high performance concrete for precast decked bulb-tee bridge girder connections
Creator:
Alkhalaf, Abdulsalam Mohammed Hassan
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Master's ( Master of science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Civil Engineering, CU Denver
Degree Disciplines:
Civil engineering
Committee Chair:
Li, Chengyu
Committee Members:
Rutz, Frederick R.
Nogueira, Carnot

Notes

Abstract:
In recent decades, the ultra-high performance concrete (UHPC) has been chosen by many state departments of transportation to be an ideal solution for many challenges of the precast bridge system. The UHPC material has numerous advantages over the conventional concrete, due mainly to the steel fiber reinforcement. The fundamental goal of this research study is to improve the design that provides satisfactory performance in a connection area between two adjacent decked bulb-tee bridge girders. Therefore, the design and building of statically determinate structures were completed to know the bending moment capacity and deflection for any applied load. The compressive strength of the UHPC material was up to 28 ksi based on the ASTM-standard test procedures in this study and the tensile strength above 2 ksi. The experimental test results of the simulated deck, which represents the decked bulb-tee girder system with UHPC material, provided higher strength than the cast-in-place deck by approximately 30 percent in terms of the bending moment capacity in this study. The specific shear key and noncontact lap-splice of the connection area show acceptable behavior where the failure mode of all simulated decks was the conventional concrete crushing.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Abdulsalam Mohammed Hassan Alkhalaf. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

EXPERIMENTAL INVESTIGATION OF ULTRA HIGH PERFORMANCE CONCRETE FOR PRECAST DECKED BULB TEE BRIDGE GIRDER CONNECTIONS by ABDULSALAM MOHAMMED HASSAN ALKHALAF B.S., Imam Abdulrahman Bin Faisal University, 2014 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering Program 2018

PAGE 2

ii © 2018 ABDULSALAM MOHAMMED HASSAN ALKHALAF ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Science degree by Abdulsalam Mohammed Hassan Alkhalaf has been approved for the Civil Engineering Program by Chengyu Li, Chair Frederick Rutz Carnot Nogueira Date: December 15, 2018

PAGE 4

iv Alkhalaf, Abdulsalam Mohammed Hassan (M.S., Civil Engineering) Experimental Investigation of Ultra High Performance Concrete for Precast Decked Bulb Tee Bridge Girder Connections Thesis directed by Associate Professor Chengyu Li ABSTRACT In recent decades, the ultra high performance concrete (UHPC) has been chosen by many state departments of transportation to be an ideal solution for many challenges of the precast bridge system. The UHPC material has numerous advantages over the conventional concrete , due mainly to the steel fiber reinforcement. The fundam ental goal of this research study is to improve the design that provides satisfactory performance in a c onnection area between two adjacent decked bulb tee bridge girders. Therefore, the design and building of statically determinate structures were completed to know the bending moment capacity and deflection for any applied load. The compressive strength of the UHPC material was up to 28 ksi based on the ASTM standard test procedures in this study and the tensile strength above 2 ksi. The experimental test results of the simulated deck , which represents the decked bulb tee girder system with UHPC material , pr ovided higher strength than the cast in place deck by approximately 30 percent in terms of the bending moment capacity in this study. The specific shear key and noncontact lap splice of the connection area show acceptable behavior where the failure mode of all simulated decks was the conventional concrete crushing. The form and content of this abstract are approved. I recommend its publication . Approved: Chengyu Li

PAGE 5

v ACKNOWLEDGEMENTS Firstly, praise to Allah for all grace that was or is being, and all tha nks to Allah, who is able to do anything, for helping me to achieve this thesis. Secondly, I am extremely thankful to Prof. Chengyu Li, who was my advisor, for giving me enough learning, motivation, enthusiasm, and patience to complete my graduate experien ce. Therefore, I really would like to invest in this opportunity to appreciate his knowledge, experience, and skills. Furthermore, I would like to express my thanks to Tom Thuis, Jac Corless, and Mohammed Edqam for helping me to prepare my samples and test s. I would also like to thank the Lafarge company , represented by Gregory Nault , for provid ing Ductal UHPC material and teaching me how to mix and cast the UHPC material. Additionally, my great thanks to Saudi Arabia, which is my country, for the full scholarship to pursue my graduate education. Finally, I would like to express my gratitude to my father , Mohammed Alkhalaf ; my mother , Dakhila Alkhalaf ; my brothers Hisham, Abduljaleel, and Shaker ; my sisters Basma and Nura ; my wife , Malak Alkhalaf ; m y child , Abduljaleel ; and all my family, who are by my side whenever I need them. I completely and gratefully acknowledge that without their love and care , I would not be what I am toda y.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ .............. 1 Overview ................................ ................................ ................................ ............................. 1 Problem Statement ................................ ................................ ................................ .............. 3 Scope ................................ ................................ ................................ ................................ ... 5 Research Objective ................................ ................................ ................................ ............. 6 II. LITERATURE REVIEW ................................ ................................ ................................ ... 7 Common UHPC Connections ................................ ................................ ............................. 7 Lap Spliced Rebar Connections with Different Strengths of UHPC Materials ............... 11 Investigation of UHPC M aterial for Longitudinal Joints in Deck ed Bulb Tee Bridge Girders by Peruchini (2017) ................................ ................................ .............................. 12 The Curing of UHPC ................................ ................................ ................................ ........ 15 Side By Si de Box Beam Bridge System vs. Decked Bulb Tee Beam Bridge System with CFRP and UHPC ................................ ................................ ................................ .............. 16 The Bond Strength Between UHPC and Reinforcing Bars ................................ .............. 19 III. EXPERIMENTAL TEST PROGRAM ................................ ................................ ............. 21 Overview ................................ ................................ ................................ ........................... 21 Mix Design ................................ ................................ ................................ ........................ 22 Overview ................................ ................................ ................................ ..................... 22 Conventional Concrete (Quikrete) ................................ ................................ .............. 22 UHPC ( Ductal ) ................................ ................................ ................................ ......... 23 Mixing and Placing Procedure ................................ ................................ .......................... 26

PAGE 7

vii Material Strength Tests ................................ ................................ ................................ ..... 30 Simulated Deck Preparation ................................ ................................ ............................. 34 Ov erview ................................ ................................ ................................ ..................... 34 Simulated Deck Design ................................ ................................ ............................... 35 Simulated Deck Formwork Preparation ................................ ................................ ...... 41 Curing Processes ................................ ................................ ................................ ......... 44 Installing the I nstrumentation ................................ ................................ ..................... 46 IV. EXPERIMENTAL RESULTS ................................ ................................ .......................... 49 Materials Test Results ................................ ................................ ................................ ....... 49 Compression Cylinders Tests ................................ ................................ ..................... 49 Flexural Beams Tests ................................ ................................ ................................ .. 57 Split tension Cylinders Tests ................................ ................................ ...................... 60 Simulated Deck Test Results ................................ ................................ ............................ 62 V. ANALYSIS AND DISCUSSION OF RESULTS ................................ ............................ 70 Material Strength ................................ ................................ ................................ .............. 70 Simulated Deck Strength ................................ ................................ ................................ .. 72 VI. CONCLUSIONS, RECOMMENDATIONS AND FUTURE RESEARCH .................... 77 Conclusions ................................ ................................ ................................ ....................... 77 Recommendations ................................ ................................ ................................ ............. 78 Future Research ................................ ................................ ................................ ................ 79 REFERENCES ................................ ................................ ................................ ............................. 81 APPENDIX A.Design and Analysis of Cast in Place Deck (No Connection) ................................ ........... 83

PAGE 8

viii Analysis o f Doubly Reinforced Beam Sections ................................ ............................... 85 SAP2000 S oftware for Cast in P lace D eck A nalysis ................................ ....................... 88 B.Additional Experimental Test Results ................................ ................................ ................. 90 Normal Concrete Material Tests ................................ ................................ ....................... 90

PAGE 9

ix LIST OF TABLES TABLE 2.1. The mechanical and durability properties of the UHPC types (Haber and Graybeal, 2018) .. 11 2.2. Results of all simulated deck specimens by Peruchini (2017) ................................ ............... 14 2.3. Compressive strength of cubes by Prem, Ramachandra, and Bharatkumar (2015) ............... 16 3.1. Mixing water for Quikrete concrete mix by Quikrete. ................................ .......................... 22 3.2. Mix design quantities for conventional co ncrete ................................ ................................ ... 23 3.3. Mechanical and durability properties of Ductal JS1000 (UHPC) (Lafarge, 2018) ............. 24 3.4. UHPC Ductal mix proportions ................................ ................................ ............................ 25 3.5. Description of UHPC material strength specimens ................................ ............................... 31 3.6. Description of conventional concrete material strength specimens ................................ ....... 31 3.7. Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) ................................ ................................ ................................ ..................... 36 3.8. The properties of the noncontact lap splice in the connection area ................................ ....... 36 4.1. Results of UHPC compression cylinder tes t at 7 days ................................ ........................... 51 4.2. Results of UHPC compression cylinder test at 14 days ................................ ......................... 51 4.3. Results of UHPC compression cylinder test at 28 days ................................ ......................... 52 4.4. Results of UHPC flexural beams ................................ ................................ ........................... 58 4.5. Results of split tension cylinders tests ................................ ................................ ................... 60 4.6. Results of three full simulated decks tests . ................................ ................................ ............ 63 5.1. The range of the UHPC material properties by FHWA. ( Russell and Graybeal, 2013) ........ 70

PAGE 10

x LIST OF FIGURES FIGURE 1.1. Precast decked bulb tee bridge girders being connected together ................................ ........... 2 1.2. Existing CDOT connection details ................................ ................................ .......................... 4 1.3. Proposed structure for UHPC connection ................................ ................................ ................ 5 2.1. Transverse joint for precast bridge deck panels connection by using UHPC material (Graybeal, 2014) ................................ ................................ ................................ ..................... 8 2.2. UHPC composite connection between deck panels and girder as developed by NYSDOT (Graybeal, 2014) ................................ ................................ ................................ .... 9 2.3. UHPC connection detail for adjacent box beam (Graybeal, 2014) ................................ ....... 10 2.4. Midspan moment displacement behavior under ultimate loadi ng (Haber and Graybeal, 2018) ................................ ................................ ................................ ................................ ..... 12 2.5. Interface crack between the UHPC and the concrete (Peruchini, 2017) ................................ 14 2.6. The development of the longitudinal deck cracks between side by side box beams ............ 17 2.7. Load deflection curves for all beams in the bridge mode l by Grace, Bebawy, and Kasabasic (2015) ................................ ................................ ................................ .................. 18 2.8. Stress slip response of the pullout test that is completed by Haber and Graybeal (2018) ..... 20 3.1. Steel fibers in UHPC ................................ ................................ ................................ .............. 26 3.2. Vibrating the fresh concrete using the concrete vibrator ................................ ....................... 27 3.3. Casting UHPC in t he connection area ................................ ................................ ................... 29 3.4. The 220K MTS machine: (a) The compression cylinders test; (b) The split tension cylinder test ................................ ................................ ................................ .......................... 32 3.5. The Forney machine for the compression cylinders test ................................ ....................... 33

PAGE 11

xi 3.6. The 20K MTS machine for the flexural strength test ................................ ............................ 33 3.7. Locati on of the positive moment on the joint area from truck wheel loading ....................... 34 3.8. Plan view of the full simulated deck specimen ................................ ................................ ...... 37 3.9. Section A A of plan view in Figure 3.8. ................................ ................................ ................ 37 3.10. Section B B of plan view in Figure 3.8. ................................ ................................ .............. 38 3.11. Section C C of plan view in Figure 3.8. ................................ ................................ .............. 38 3.12. Transverse Joint (connection). ................................ ................................ ............................. 40 3.13. Formwork preparation for casting normal concrete: (a) Shear key side of the formwork; (b) Wood parts were being covered by transparent tape ................................ ...................... 41 3.14. Completed wood formwork for casting on e normal concrete panel ................................ .... 42 3.15. Completed wood formwork for casting UHPC material in connection area ....................... 43 3.16. Reinforcement measurements before casting the UHPC material (3 in. center to center) .. 43 3.17. Curing the normal concrete panels by concrete blank et ................................ ...................... 44 3.18. Curing the UHPC joints by multiple plastic sheets ................................ ............................. 45 3.19. Longitudinal section view of the simulated deck test setup ................................ ................ 47 3.20. Plan view of the simulated deck test setup ................................ ................................ .......... 47 3.21. Computer is being connected for the simulated deck test ................................ .................... 48 4.1. T he first UHPC compression cylinder at 7 days after testing (1 C UHPC 7d) ..................... 53 4.2. The second UHPC compression cylinder at 7 days after testing (2 C UHPC 7d) ................ 53 4.3. The third UHPC compression cylinder at 7 days after testing (3 C UHPC 7d) .................... 53 4.4. The first UHPC compression cylinder at 14 days after testing (1 C UHPC 14d): (a) ................................ ................................ ......... 54

PAGE 12

xii 4.5. The second UHPC compression cylinder at 14 days after testing (2 C UHPC 14d): (a) ................................ ................................ ......... 54 4.6. T he third UHPC compression cylinder at 14 days after testing (3 C UHPC 14d): (a) ................................ ................................ ......... 54 4.7. The first UHPC compression cylinder at 28 days after testing (1 C UHPC 28d): (a) ................................ ................................ ......... 5 5 4.8. The second UHPC compression cylinder at 28 days after testing (2 C UHPC 28d): (a) ................................ ................................ ......... 55 4.9. The third UHPC compression cylinder at 28 days after testing (3 C UHPC 28d): (a) screen ................................ ................................ ......... 55 4.10. The fourth UHPC compression cylinder at 28 days after testing (4 C UHPC 28d): (a) ................................ ................................ ......... 56 4.11. The fifth UHPC compression cylinder at 28 days after testing (5 C UHPC 28d): (a) Failure shape; (b) Forne ................................ ................................ ......... 56 4.12. The sixth UHPC compression cylinder at 28 days after testing (6 C UHPC 28d): (a) ................................ ................................ ......... 56 4.13. The failure shape of each UHPC flexural beam ................................ ................................ .. 59 4.14. The failure shape of each UHPC split tens ion cylinder ................................ ....................... 61 4.15. Load vs. Mid Span displacement diagram for SD 1 ................................ ........................... 64 4.16. Load vs. Mid Span displacement diagram for SD 2 ................................ ........................... 64 4.17. Load vs. Mid Span displacement diagram for SD 3 ................................ ........................... 65 4.18. Load vs. Strain diagram for SD 1 ................................ ................................ ........................ 65 4.19. Load vs. Strain diagram for SD 2 ................................ ................................ ........................ 66

PAGE 13

xiii 4.20. Load vs. Strain diagram for SD 3 ................................ ................................ ........................ 66 4.21. Moment vs. PI gauge for SD 1 ................................ ................................ ............................ 67 4.22. Moment vs. PI gauge for SD 2 ................................ ................................ ............................ 67 4.23. Moment vs. PI gauge for SD 3 ................................ ................................ ............................ 68 4.24. The failure mode of the normal concrete being crushed in SD 1 ................................ ........ 68 4.25. The failure mode of the normal concrete being crushed in SD 2 ................................ ........ 69 4.26. The failure mode of the normal concrete being crushed in SD 3 ................................ ........ 69 5.1. The bending moment vs. displacement for all simulated decks ................................ ............ 73 5.2. Destroyed shape inside the UHPC joint of SD 1 ................................ ................................ ... 75 5 .3. Destroyed shape inside the UHPC joint of SD 2 ................................ ................................ ... 75 5.4. Destroyed shape inside the UHPC joint of SD 3 ................................ ................................ ... 76

PAGE 14

1 CHAPTER I INTRODUCTION Overview Using a traditional construction process such as concrete mixing truck s is difficult in bridge projects that are constructed in the remote and rural area s throughout the s tate of Colorado . Consequently , the civil engineers desire to discover an alternative method , which is precast concrete. This option of construction process serves to eliminate some cast in place concrete constructions . Precast concrete is a construction invention that uses a mold or a form for casting concrete in the factory . The precast concrete is then transported to the construction site and lifted into place by a crane after having been cured in a controlled environment. One of the most common types of precast concrete used in bri dge project s is the precast girder. In order to create a driving surface, a cast in place concrete deck will be placed on the top of the precast girders. Although use of the precast concrete approach is faster than a cast in place approach, the site must s till be able to accommodate the concrete truck for pouring the driving deck. T he Colorado Department of Transportation (CDOT) has been conducting research on the decked bulb tee girder , which is on e type of precast bridge girder . CDOT is working to develop the us e of precast girders and rejecting the cast in place construction . In the precast factory, decked bulb tee girder s can be manufactured in any grouping of various depths ( from 33 in. to 72 in. ) and various flange widths ( from 48 in. to 96 in. ). In these girders, the cast in place concrete for the driving deck can be replaced with precast monolithically with the girders. Therefore, after lifting the girders in the site (see F igure 1.1 . ), only the connection between the

PAGE 15

2 adjacent girder flange s must be achieved. This type of construction is able to reduce the labor, onsite overcrowding, and time, which serves to reduce The first ultra high performance concrete ( UHPC) bridge is a pedestrian bridge that was constructed in Canada in 1997 . Canada now has at least 26 bridges built with UHPC. In approximately 2000, the UHPC became available in the United States marketplace, and the Federal Highway Administration (FHWA) started investigating the utiliz ation of UHPC for h ighway structures in 20 01. The UHPC materials often consist of a fine sand, Portland cement, silica fume, steel fibers, high range water reducing admixture (HRWR), and water. Based on the materials composition , the UHPC is one type of fiber reinforced concrete that has a better compression, tension, ductility, and bond strength than conventional concrete. Generally, this composite Figure 1. 1 . Precast decked bulb tee bridge girders being connected together

PAGE 16

3 material with discontinued steel fibers can provide compressive strength above , and tensile strength above . Lafarge company in Chicago developed its own UHPC called Ductal , and this material governs the United States marketplace. Many experiments have been conducted using Ductal material, and the most extensive investigations have been accomplished by Benjamin Graybeal at FHWA (Graybeal, Design and Construction s of F iel d Cast UHPC Connection 2014). After submitting the proposal about this project to Lafarge, they were willing to be a provider in this study ; h ence, the UHPC that has been used in this study is Du ctal . A disadvantage of using UHPC is its cost. For any construction project currently planning to use UHPC, Ductal is only the UHPC product in the market. This led to the stability of the UHPC cost because there is no other provider that is commercial ly competit ive . The existence of more suppliers for UHPC is going to reduce the cost, and this will happen when the avai lability of UHPC exists in the market. T he availability of UHPC in the marketplace is the second challenge, which also has an effect on the cost. For these reasons, Ductal is approximately 20 to 30 times the cost of conventional concrete , mainly due to t he high quality of UHPC. Problem Statement The design method to connect the two adjacent flanges in the decked bulb tee girders is the critical point for this type of precast currently. The CDOT has been studying some of these types of precast connections, and one of them is what Figure 1.2 . shows. In F igure 1.2 . , the thickness of the panels is 9 in . and the joint area width is 6 in. The panels will be placed at a 1/4 in. space between the panels from the bottom, and this gap can vary due to tolerances of the panels. Typically, the cambers of adjacent panels will not be identical, and jacking up or down will be needed to achieve the proper alignment. After placing the grout materials for a joint area,

PAGE 17

4 the vibration is particularly important to ensure the filling of all voids in the joint area. Grinding with 1/8 in. on the top of the joint area is mandatory to make a smooth driving surface. Even though the minimum thickness required by the CDOT for the deck is 8 in., they would know the results of 6 in. thickness for this type of deck ed bulb tee bridge girder connection with UHPC material. In fact, the connection between deck ed bulb tee bridge girders has shown damage when subjected to heavy truck loading. Generally, this damage is shown by the spalling of the grout materials surrounding the reinforcement in the joint area. As a result of this obvious damage , the precast decked bulb tee bridge girders are utilized , essentially , on streets with light traffic. CDOT proposed to study materials other than conventional concrete with the goal of using precast decked bulb tee bridge girders on the heavy traffic road, and they decided to investigate the UHPC material. Figure 1. 2 . Existing CDOT connection details

PAGE 18

5 Scope Figure 1.3 . shows the proposed structure that has been used in this study. As CDOT and Lafarge are especially interested in UHPC purposes in decked bulb tee bridge girder connections, each one wanted to investigate specific things in this study. Therefore, CDOT needs to investigate the UHPC material with this specific transverse joint shape shown in Figure 1.3. For Lafarge, they need to study , also , this transverse joint shape, but with the overlap noncontacted reinforcement. Furthermore, Lafarge wants no space between two adjacent decks from the bottom before casting UHPC in order to avoid the formwork for casting UHPC material. Figure 1. 3 . Proposed structure for UHPC connection Plan View Transverse Join t

PAGE 19

6 Research Objective The monotonic ultimate load and the vertical deflection at midspan of full simulated deck specimen s , which represents the connection between the decked bulb tee bridge girders, are the primary goals of this developed experiment. These two values can achieve a comparison between using UHPC materials to complete the connection between two adjacent precast decked bulb tee bridge girders that was investigated experimentally in this stud y and the cast in place deck (no connection).

PAGE 20

7 CHAPTER II LITERATURE REVIEW Common UHPC Connections T he advantages of the UHPC materials and durability properties are able to modify and improve the connection details for prefabricated elements. Graybeal (2014) wrote that UHPC has high compressive and post cracking tensile strengths , which are great properties, but UHPC has steel fiber reinforcement and discontinuous pore structure that provide additional material property benefits. This explains that not only the high compressive and post cracking tensile strengths are the benefits of UHPC, but also the components of UHPC contribute to these advantages to afford an internal distribution of stresses, reduced rebar development and splice length, and better bond strength. F iel d Cast UHPC Lap Joint Between Adjacent Precast Bridge Deck Pane ls The first common connection is a field cast UHPC lap joint between adjacent precast bridge deck panels. In this kind of connection, the rebar of each adjacent panel will overlap in the joint area, and this area will be filled with UHPC. Graybeal (2014) reported that this type of connection is going to use short, straight rebar at typical spaces. Therefore, using UHPC in this type of connection c an show easier construction method s than the conventional concrete . Figure 2.1 . illustrates one example of the connection between adjacent precast bridge deck panels by using UHPC material. These details are for reconstruction of a bridge on County Road 47 over Trout Brook near Stockholm, NY. The top and bottom mats are lap spliced with No. 5 epo xy coated reinforcing bars spaced at 12 in. The thickness of the deck panels is 8.5 in. The joint area in this example is specified to be 8 in. wide and the lap splice is a minimum of

PAGE 21

8 6 in. R educing the volume of the UHPC material will require a simple formwork to make the connection. Composite Action Between Precast Deck Panel and Girder by Using UHPC The second common connection is for composite action between the precast deck panel s and girder using UHPC. Shear studs or extensional reinforcing bars from the girder into a block out pocket in the deck, which is filled with grout, are used in order to create a composite girder to a deck in the conventional design. Thus, using UHPC material can improve the conventional b l ock out pocket design. Graybeal (2014) reported the conventional design requires deck overlays over the precast deck panels to prevent the water ingress through the joint between the prefabricated deck panels and the shear stud pocket. As a result of additional construction activities to create deck ov erlays Figure 2. 1 . Transverse joint for precast bridge deck panels connection by using UHPC material (Graybeal, 2014)

PAGE 22

9 in the conventional design, using UHPC in this case is able to eliminate the deck overlays because this material has a great bond strength for joining precast surfaces . Also, the liquid ingress through the joint between the prefabricated deck panels will be eliminated due to the discontinuous pore structure of UHPC materials. Figure 2.2 . shows an example of a composite connection between a precast deck panel and girder by using UHPC. This detail is developed for a highway bridge near Syracuse in New York state. This UHPC connection combines the panel to panel connection and the deck to girder connection that takes place along the girder line. Connection Between Adjacent Box Beams and Other Longitudinal Elements by Using UHPC Material The third common connection is for adjacent box beams and other longitudinal elements by using UHPC material. Graybeal (2014) wrote that using conventional connection details in this case can often show destitute performance such as various deflection between beams, water ingress, and overlay cracking , which all lead to the failure of the connection. This means that the Figure 2. 2 . UHPC composite connection between deck panels and girder as developed by NYSDOT (Graybeal, 2014)

PAGE 23

10 discovery of a more beneficial grout material is very important, and UHPC material could cover all these problems. In the U nited State s , the first box beam bridge to use UHPC material is the Sollars Road Bridge over Lees Creek in Fayette County, OH. The span bridge measures 61 ft . and comprises seven adjacent box beams. Figure 2.3. shows the connection detail for adjacent box beams. Also in this connection, the reinforcing bars will lap within the joint area. Figure 2. 3 . UHPC connection detail for adjacent box beam (Graybeal, 2014)

PAGE 24

11 Lap Spliced Rebar Connections with Different Strengths of UHPC Materials Haber and Graybeal (2018) conducted an experimental study on five different types of UHPC materials. The main goal of this research was to study the behavior of the connection between two adjacent prefabricated bridge decks . Table 2.1 . lists the mechanical and durability properties of the UHPC material types which were used by Haber and Graybeal (2018). Figure 2.4. shows the relationship between the applied moment and the midspan displacement for all the UHPC types unde r the monotonic ultimate loading that is controlled by a displacement rate of 1.3 mm/min. Haber and Graybeal reported that they have drawn each curve shown in Figure 2.4. at the point of the peak load, which correspond s to the concrete crushing moment. Als o, they mentioned that the failure of each specimen was due to the crushing of the normal concrete. Test Material type U A U B U C U D U E 7 day compressive strength (MPa) 120 110 122 137 110 Direct tension strength (MPa) 6.28 8.35 5.55 7.87 6.88 Bond strength to precast concrete (MPa) 2.07 3.45 2.20 2.55 2.62 0.15 0.15 0.16 N/A 0.17 Static flow (mm) 146 254 102 191 181 Dynamic flow (mm) 219 254 194 229 222 Table 2. 1 . The mechanical and durability properties of the UHPC types ( Haber and Graybeal, 2018 )

PAGE 25

12 Investigation of UHPC M aterial for Longitudinal Joints in Deck ed Bulb Tee Bridge Girders by Peruchini (2017) Peruchini (2017) studied the specific UHPC mix that is developed by Washington State University. The goal of Peruchini (2017) experimental study was to investigate the structu ral performance of this specific UHPC mix in a reinforced spliced connection for concrete deck ed bulb tee bridge girders. Figure 2. 4 . Mid span moment displacement behavior under ultimate loading (Haber and Graybeal, 2018)

PAGE 26

13 Also, the important point was to study the bond strength of epoxy coated reinforced bars in the connection area with different joint widths. Meanwhile, it was important to consider the tensile strength of the UHPC when the reinforced bars in the connection are stressed axially in tension. Peruchini (2017) followed the ASTM standard in his study for material tests. The results were the UHPC compressive strength was up to 16 ksi at 14 days , the UHPC tension strength above 2 ksi, and the bo n d strength equal t o 7 ksi. Table 2.2. lists the results of all simulated deck specimen s of Peruchini (2017) experimental study. It is clear that the bending moment increases when the UHPC connection width increases. That maximum bending moment was 562.2 kip in. However, the cover can cause differences in the bending moment values. As shown in T able 2.2. , S pecimen # 7 has a 5 in. joint width, 1.75 in. cover , and achieved 487.4 kip in bending moment , which is higher than S pecimen # 6 that has a 474.1 kip in bending moment, 6 in. joint width, and 1 in. cover. Peruchini (2017) wrote that there was no flexural crack in the body of the UHPC. Consequently, the failure mode was bond failure between the UHPC material and the conventional concrete as shown in Figure 2.5. In fact, this failure mod e was obtained for all simulate d deck s in Peruchini .

PAGE 27

14 Specimen # Joint Width/Cover/Offset (in) Interfacial Bending moment (K in) 1 7/1/0 562.2 2 3/1/0 245.8 3 5/1/0 391.7 4 5/1/2 336.9 5 5/1/1 400.6 6 6/1/0 474.1 7 5/1.75/0 487.4 8 6/1.75/0 471.5 Table 2. 2 . Results of all simulated deck specimens by Peruchini (2017) Figure 2. 5 . Interface crack between the UHPC and the concrete (Peruchini, 2017)

PAGE 28

15 The Curing o f UHPC The curing process for any type of concrete is extremely important. Achievement of the highest compressive strength of concrete depends on how the samples are going to be treated. As there are many types of concrete curing , such as water curing, heat curing, and stea m curing, the highest compressive strength of UHPC could be achieved with heat curing. Prem, Ramachandra, and Bharatkumar ( 2015 ) reported that the heat curing is the best technique for UHPC material, which showed the highest compressive strength of 28.42 ksi (196 MPa) at 28 days in their experiment. Due to the combination of an enormous amount of quartz plus to portlandite and tricalcium alumino ferrite (brownmillerite), the high compressive strength could be attained with heat curing. Therefore, the stren gth of UHPC can be increased by controlled increases in temperature and the time of curing. By using heat curing, the hydration mechanism will be different ; thus , the reaction of the ingredients is going to increase and lead to the high strength gain of UHPC. Table 2.3. ( Pr em, Ramachandra, and Bharatkumar , 2015 ) lists the compressive strength of cubes that have been cured with water, steam, and heat. At the age of 7 days, s team curing and heat curing created better compressive strength of 120 MPa and 160 MPa , respectively , t han water curing of 64 MPa. The heat curing could achieve the highest compressive strength in all stages after 3 days, and the minimum percentage was 25 percent more than steam curing at the age of 7 days. At the age s of 28 days and 14 days, there were not many differences between water curing and steam curing, but the steam curing reached the compressive strength of 50 percent , approximately , more than water curing at the age of 7 days.

PAGE 29

16 Side By Side Box Beam Bridge System vs. Decked Bulb T ee Beam Bridge System with CFRP and UHPC Since the 1950s, the side by side box beam bridge system has been used widely, and it is one of the desired precast prestressed bridge systems. The side by side b ox beam bridge system has many great properties , such as shallow depth, unnecessary formwork on site, fast construction process, and considerable torsion capacity. However, the problem of the development of the longitudinal deck cracks between the box beam s began in the last few decades. These deck cracks can lead to the accelerated deterioration of the superstructures due to the seeping through the cracks into the sides of the box beams ( s ee F igure 2. 6 . ). Also , the full inspection and maintenance are difficult processes for this crack problem because there is no space between the box beams. Age at testing: d Compressive strength: M Pa Water curing Steam curing Heat curing 3 53 53 53 7 64 120 160 14 111 128 185 28 144 142 196 Table 2. 3 . Compressive strength of cubes by Prem, Ramachandra, and Bharatkumar (2015)

PAGE 30

17 Therefore, a decked bulb tee beam bridge system is a potential alternative to a side by side box beam bridge system. Consequently, Grace, Bebawy, and Kasabasic (2015) decided to investigate the decked bulb tee beam bridge system reinforced with carbon fibe r reinforced polymer (CFRP) materials instead of the conventional steel reinforcement. Also, UHPC has been used to finish the connection between the beams instead of the conventional concrete. Use of UHPC in the connection beam was for minimizing the poten tial of longitudinal deck crack ing . Thus, they conducted an experimental investigation of a complete bridge model consist ing of five beams. This bridge mode l could reach to the maximum load carrying capacity of 220 kips. As Grace, Bebawy, and Kasabasic (20 15) reported, the over stressed joint area with a clear sign of separation failure next to the intermediate beam was apparent visually. Figure 2. 6 . The development of the longitudinal deck cracks between side by side box beams

PAGE 31

18 Figure 2.7. i llustrates that the deflection of all five beams in this bridge model was equal until the bridge reached its maximum capacity of 220 kips. After the bridge reached its maximum load, the loaded beam had its different deflection curve of all other beams curv es, and this contributed to the separation failure of the joint area adjacent to the loaded beam. The experimental investigation and numerical analysis that was conducted by Grace, Bebawy, and Kasabasic (2015) clarif ied that the d ecked bulb tee bridge girder system with CFRP and UHPC is an excellent alternative to side by side box beams. The full inspection and maintenance of this type of bridge project can be easy due to there being enough space between the beams. Figure 2. 7 . Load deflection curves for all beams in the bridge model by Grace, Bebawy, and Kasabasic (2015)

PAGE 32

19 The Bond Strength Between UHPC a nd Reinforcing Bars Yuan and Graybe al (2014) reported several important points about the bond strength of reinforcing steel in the UHPC. Firstly, the bond strength can be increased by increasing the embedment length of the steel bars. This relationship is nearly linear, so the UHPC offers i mproved performance over the conventional concrete. Secondly, when the side cover increases, the bond strength increases. Thirdly, the contact lap splice samples exhibit lower bond strength than the noncontact lap splice samples, and this is probably because of the decreased contact area between the UHPC and the bars. Meanwhile, the ability of steel fiber to enhance the resistance of the UHPC will decrease due to the tight spacing between the reinforcing bars. Also, if the reinforcing bar spac ing is very large , it will decrease the bond strength due to the diagonal crack between the pullout force and the adjacent bar. Fourthly, the bond strength can be increased by increasing the compressive strength of the UHPC. Fifthly, uncoated reinforcing b ars show higher bond strength than the epoxy coated reinforcing bars. Additionally, the fiber volume fraction of the UHPC is able to influence the bond strength of reinforcing bars embedded in the UHPC. Haber and Graybeal ( 2018) conducted a pullout test t o investigate the performance of the bond strength between the UHPC with different fiber volume fractions and the embedded steel bars. They discovered that when the fiber volume increase s , the bond strength will increase. Furthermore, if the UHPC material has the fiber volume fractions less than 2 percent, the embedded bars may not have sufficient stress prior to lap sp l ice failure. As a result of this, the ductility of the UHPC connection will be reduced where the failure mode will be a lap splice failure.

PAGE 33

20 Figure 2.8. shows the comparison of the stress slip response of the pullout test that is completed by Haber and Graybeal ( 2018) . In this figure, the UHPC mixture was tested with different fiber volume fractions. It is clear that there is an increasing re lationship between the fiber volume fraction and the bond strength between the UHPC and the reinforcing bars. Figure 2. 8 . Stress slip response of the pullout test that is completed by Haber and Graybeal (2018)

PAGE 34

21 CHAPTER III EXPERIMENTAL TEST PROGRAM Overview This study is a program of three separate phases that were a series of physical experiments. In fact, all phases determined the strength of the UHPC in a noncontact lap splice connection between two adjacent precast decked bulb tee bridge girders. The program phases were as follows: 1 ) Material Mixing 2 ) Material Strength Tests 3 ) Simulated Deck Tests The experimental program studied the behavior and the ultimate load capacity of three full specimens , which represent the connection between two precast decked bulb tee bridge girders under monotonic loading. The experimental program recorded and summarized load carrying capacity, deflection, and failure modes. The primary focus of the experimental program was the simulat ed deck experiments. All the simulate d decks have the same type of the UHPC materials , which is a product from Lafarge known as Ductal . The test of each simulated deck was conducted after approximately 28 days of UHPC curing. This was completed to discover the full compression strength that is the most important feature of UHPC material in the connection between two adjacent precast decked bulb tee bridge girders.

PAGE 35

22 Mix Design Overview Two types of premix concrete materials were used in this experimen tal program. The first material is the 4000 psi conventional concrete that is provided by Quikrete ; t he second material is a UHPC known as Ductal that is provided by Lafarge. Conventional Concrete (Quikrete) The Quikrete concrete mix consists of Portland cement, sand, and gravel or stone. Actually, the Quikrete concrete mix is a pre blended mixture , so just adding water is what it needs. Table 3.1 . lists the water content for each size of Quikrete package. As an 80 lb. Quikrete concrete bag is what this experimental program utilizes . T he added water quantity was determined to be 0.925 gal for each mixing bag (Table 3.2 . ). The choosing of 0.925 gal of water for each 80 lb. concrete bag is to make sure that all the specimen s in this experimental program will have similar mix design proportions for conventional concrete. Bag size (lb) Starting water content, ( gal ) Final water content ( gal ) 80 0.739 0.739 1.136 60 0.502 0.502 0.872 40 0.370 0.370 0.555 Table 3. 1 . Mixing water for Quikrete concrete mix by Quikrete Table 3. 2 . Mix design quantities for c onventional concrete Table 3. 3 . Mixing water for Quikrete concrete mix by Quikrete. Table 3. 4 . Mix design quantities for c onventional concrete Table 3. 5 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 6 . Mix design quantities for c onventional concrete

PAGE 36

23 UHPC (Ductal ) The specific name of the UHPC material which was used in this experimental program is Ductal JS1000 that is provided by Lafarge. Ductal JS1000 is defined as the UHPC that offers higher strength, durability, ductility, and bond capacity over the tradition al concrete and high performance concrete. Ductal JS1000 is internally reinforced with discrete steel fibers that are 2 percent by volume. The primary use of Ductal JS1000 is to fill the connections between the prefabricated structural elements on site. Strength, enhanced bond, and fluidity are some of the remarkable properties of Ductal JS1000; because of this, it became the ideal product for this application. Ductal JS1000 was able to improve the design of precast structures. Table 3.3. lists the Duct al JS1000 mechanical and durability properties. The results in this table are in accordance with ASTM C1856/C1856M and curing conditions at and 50 percent R.H. Therefore, the results of Ductal JS1000 in the field may differ due to factors such as curing condition, temperature, equipment used, and mixing/testing method. Number of 80 lb. Qu ikrete concrete bags Water 1 0.925 gal Table 3. 2 . Mix design quantities for c onventional concrete Table 3. 28 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 29 . Mix design quantities for c onventional concrete Table 3. 30 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 31 . UHPC Ductal mix proportions Table 3. 32 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 33 . Mix design quantities for c onventional concrete Table 3. 34 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 35 . Mix design quantities for c onventional concrete Table 3. 36 . Mechanical and durab ility properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 37 . UHPC Ductal mix proportions Table 3. 38 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 39 . UHPC Ductal mix proportions Figure 3. 1 . Steel f ibers in UHPC

PAGE 37

24 The measurement of 4 days or less is typical when the curing temperature is more than 60 , so an accelerating admixture is required for colder temperature. 3 in. x 6 in. cylinders are used to complete the compression test s. This part measures the sustainable, post cracking, direct tensile strength of a mix that has 2 percent steel fiber by volume . For this experimental program, the Ductal JS1000 is received from Lafarge as the following three components: Property Value Age Density 150 160 lb/ft 3 Flow 7 to 10 in. diameter without visible signs of the fiber segregation Working / Set time Approx. 120 min / 15 to 20 hrs Compressive Strength At 4 days Compressive Strength At 28 days Tensile Strength At 28 days Modulus of Elasticity At 28 days Long term Shrinkage microstrain At 28 days Chloride Ion Penetrability coulombs (very low) At 56 days Freeze Thaw Resistance At 300 cycles Table 3. 3 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 53 . UHPC Ductal mix proportions Table 3. 54 . Mechanical and durability properties of Ductal J S1000 (UHPC) ( Lafarge , 2018) Table 3. 55 . UHPC Ductal mix proportions Figure 3. 8 . Steel f ibers in UHPC Figure 3. 9 . Vibrating the fresh concrete using the concrete vibrator Figure 3. 10 . Steel f ibers in UHPC Figure 3.2. Vibrating the fresh concret e using the concrete vibrator Figure 3. 11 . Vibrating the fresh concrete using the concrete vibrator Figure 3. 12 . Steel f ibers in UHPC

PAGE 38

25 1 ) Premix (dark grey): which consists of pre blended cement, sand, ground quartz, and silica fume. This premix (dark grey) was received in 50 lb. bags. 2 ) Premia150: which is super plasticizer. The definition of super plasticizer is a chemical component that is used where the suspension of dispersed particle s is required. The Premia150 was received in 5 gal pails . 3 ) Steel fibers: which has dimensions of 0.008 in. diameter and 0.5 in . long ( see Figure 3.1 . ). The tensile strength of this steel fiber is more than 290 ksi . The steel fibers were received in 44 lb . bags. Table 3.4. lists the mix proportions that are used in this experimental progr am for Ductal UHPC. The batch quantities listed in this table can finish 0.365 cubic feet of UHPC material . Ductal JS1000 (2% Steel Fiber) Components Batch Quantities (Ibs) Dark Grey Premix 50 Water 2.620 Premia 150 0.683 Steel Fiber (2%) 3.554 Table 3. 4 . UHPC Ductal mix proportions Figure 3. 52 . Steel f ibers in UHPC Figure 3. 53 . Vibrating the fresh concrete using the concrete vibrator Figure 3. 54 . Steel f ibers in UHPC Figure 3.2. Vibrating the fresh concrete using the concrete vibrator

PAGE 39

26 Mix ing and Placing Procedure Overview As this experimental program requires the mixing of two different type s of concrete materials -which are conventional concrete and UHPC -this section will explain two different mix ing procedures. The mixing procedure is not one of the main goals of this study, but it was especially important to determine the best mix ing procedure for each batch d epending on the equipment and materials that are available. Mix ing and Placing Procedure for Conventional Concrete The following steps are intended to explain the mixing and placing proce dures for conventional concrete. 1 . Measure 1.850 gal fresh water in two buckets, so each bucket has 0.925 gal . 2 . Roll the concrete bag on a hard surface to ensure the material is fresh and does not include any lump s . Figure 3.1. Steel f ibers in UHPC Figure 3. 148 . Vibrating the fresh concrete using the concrete vibrator Figure 3. 149 . Steel f ibers in UHPC Figure 3.2. Vibrating the fresh concrete using the concrete vibrator Figure 3. 150 . Vibrating the fresh concrete using the concrete vibrator Figure 3. 151 . Steel f ibers in UHPC

PAGE 40

27 3 . As two 80 lb. bags of Quikrete concrete a re mixed at the same time, the two 80 lb. bags a re p oured in to the mixer . 4 . Turn on the mixer and mix for 2 to 3 minutes to make sure that the concrete does not have any lump s . 5 . During the mixing of the concrete, start adding the water gradually . 6 . Mix for 5 to 8 minutes before turning off the mixer, and make sure that all concrete mixture is wet . 7 . Pour the fresh concrete in the wheelbarrow and start filling the specimen . 8 . As the vibration is required to make sure there are no air bubbles and aggregate lumps, the concrete vibrator is used (see Figure 3.2 . ) . 9 . Use the steel finishing trowels to smooth the surface . Figure 3.2. Vibrating the fresh concrete using the concrete vibrator Figure 3. 227 . Vibrating the fresh concrete using the concrete vibrator

PAGE 41

28 Mix ing and Placing Procedure for UHPC The following steps are intended to explain the mixing and placin g procedures for UHPC materials. 1 . As two 50 lb. bags of Ductal premix a re mixed at the same time, measure 5.24 lbs. of fresh water in one bucket, and weigh 1.366 lbs. of Premia 150 (super plasticizer) in another bucket. 2 . Weigh 7.108 lbs. of s teel fiber in one bucket. 3 . Roll the Ductal premix bag on a hard surface to ensure the material is fresh and does not include any lump s . 4 . Turn on the mixer and start pouring the premix bags one by one ; then mixing the Ductal pre mix for a duration of 3 minutes to make sure to homogenize all components in the Ductal premix. As the UHPC material needs a strong mixer, the pan mixer was used instead of the drum mixer that was used for the conventional concrete. 5 . Start adding the liqu ids. Gradually add the Premia 150 (super plasticizer) in first , and follow with the water to remove any super plasticizer that is sticking to the mixer parts. This step should be finished into the batch within 2 minutes. 6 . After Step 5, mix at least for 3 minutes. I t will then be apparent that the wetted mixture will change from the fine granular mix to the semi plastic state. At this point, start adding the steel fiber gradually into the mixture . Large balls of steel fiber may form if the steel fiber addition is poured too fast and not p o ured gradual ly . This step took approximately 8 minutes.

PAGE 42

29 7 . Discharge the mixer into a bucket after 2 to 3 minutes from S tep 6 to ensure that the steel fibers are carefully dispersed. The total time for on e mixing procedure of two 50 lb. bags of Ductal premix was approximately 18 to 20 minutes. 8 . Fill the specimen with UHPC from the bucket and do not use any vibrators. The vibration may cause steel fiber segregation ( s ee F igure 3.3 . ). Figure 3.3. Casting UHPC in the connection area Table 3. 199 . Description of UHPC material strength specimens Figure 3. 291 . Casting UHPC in the connection area

PAGE 43

30 Material Strength Tests This section explains the testing methods that this experimental program followed to determine the strength of the materials that were used in this study. These tests are completed to determine the compression and flexural strength of the specific mix of UHPC ( Ductal JS1000) and conventional concrete (Quikrete). All the compression tests were completed by compression cylinders that have dimension. Also, all the flexural strength tests were finished by prisms that have dimension s of All these prisms have a notch in the middle of the span with dimensions of half of the height and all the way through the width of the prisms. Table 3.5. lists the detail of UHPC material tests, and Table 3.6. lists the detail of conventional concrete tests. Two machines were used to complete the compression cylinders tests. The first machine was the 220k MTS machine ( s ee F igure 3.4 . ). This machine was used for all conventional concrete cylinders, the UHPC cylinders at 7 days, and the split tension cylinder test. However, the 220K MTS machine was not suitable for UHPC compression cylinders at 14 and 28 days because the expected ultimate load was more than 220 kips. Consequently, it was necessary to find an alternative option , which is the Forney machine ( s ee F igure 3.5 . ). The 20K MTS machine was used for all flexural strength tests. With this machine, the prism was being tested under three point loads ( s ee F igure 3.6 . ).

PAGE 44

31 Material Test Specimen Name Age (days) Number of Specimen Machine for Testing Compression Cylinders 7 3 220K MTS 14 3 Forney 28 6 Forney Flexural Beams 73 6 20K MTS Split tension cylinders 75 3 220K MTS Material Test Specimen Name Age (days) Number of Specimen Machine for Testing Compression Cylinders 7 3 220K MTS 28 6 220K MTS Flexural Beams 80 3 20K MTS Table 3. 5 . Description of UHPC material strength specimens Table 3. 359 . Description of conventional concrete material strength specimens Table 3. 360 . Description of UHPC material strength specimens Table 3. 361 . Description of conventional concrete material strength specimens Table 3. 362 . Description of conventional concrete material strength specimens Table 3. 363 . Description of UHPC material strength specimens Table 3 . 364 . Description of conventional concrete material strength specimens Table 3. 365 . Description of UHPC material strength specimens Table 3. 366 . Description of conventional concrete material strength specimens Table 3. 6 . Description of conventional concrete material strength specimens Table 3. 470 . Description of conventional concrete material strength specimens Table 3. 471 . Description of conventional concrete mat erial strength specimens Table 3. 472 . Description of conventional concrete material strength specimens

PAGE 45

32 Figure 3.4. The 220K MTS machine : (a) The compression cylinders test; (b) The split tension cylinder test Figure 3. 355 . The 220K MTS machine : (a) The compression cylinders test; (b) The split (a) Fi gu re 3. 3 71 . Th e Fo rne y ma chi ne for the (b) Fi gu re 3. 3 07 . Th

PAGE 46

33 Figure 3.5. The Forney machine for the compression cylinders test Figure 3. 487 . The 20K MTS machine for the flexural strength test Figure 3. 488 . The Forney machine for the compression cylinders test Figure 3.6. The 20K MTS machine for the flexural strength test Figure 3. 489 . Location of the positive moment on the joint area from truck wheel loading Figure 3. 490 . The 20K MTS machine for the flexural strength test Figure 3.6. The 20K MTS machine for the flexural strength test

PAGE 47

34 Simulated Deck Preparation Overview In order to investigate the performance of the UHPC materials connecting between two adjacent bulb tee bridge girder decks, these experiments tested a region of a bridge deck exposed to a posi tive bending moment due to a single maximum truck wheel loading placed on the center of the connection area between two precast bulb tee girder decks, as demonstrated by F igure 3.7 . Figure 3.7. Location of the positive moment on the joint area from truck wheel loading Table 3. 509 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Figure 3.7. Location of the positive moment on the joint area from truck wheel loading

PAGE 48

35 Building and testing of statically determinate specimens focusing on the joint area will be acceptable since the bending moment is approximately constant over a short distance of the connection area. So , confidently, the resulted moments can be determined and compared with the demands. Simulated Deck Design Figure s 3.8 . through Figure 3.11 . illustrate the geomet ric detail of the precast deck ed bulb tee connection specimen. The purpose of the simulated dec k design is to represent the deck ed bulb tee bridge girder system that is used in the field currently. As illustrated i n F igure 3.8 . , t he simulated deck specimen will have two separate longitudinal reinforced concrete panels. The first panel includes eight #5 epoxy coated bars ; four in the compression layer, four in the tension layer. The second panel will have six #5 epoxy coated bars ; thre e in the compression layer, three in the tension layer. Therefore, the bars of two panels will overlap in the noncontact spliced configuration joined by UHPC. Based on this, the panels will be 2 f t. wide as required to fit the #5 rebar spaced at 3 in. insi de the UHPC joint. Design of the lap splice The design of the lap splice was according to the guidance that was published by the U.S. Federal Highway Administration (FHWA) (Graybeal, 2014). The clear cover was an exception to this guidance, and the reason is to investigate the worst case scenario. Table 3.7 . (Graybeal, 2014) shows the minimum clear cover and embedment length of reinforcing bars in the UHPC material.

PAGE 49

36 In order to use T able 3 . 7. , the following conditions must be met: field cast UHPC with 2 percent high strength steel fiber reinforcement by volume and a minimum compressive strength of 14 ksi . As Ductal material used in this study meets these conditions, T able 3.7. can be used to illustrate the lap splice design of the precast deck ed bulb tee bridge girder connection s . Table 3.8. lists the noncontact lap splice details in the joint area for this study (see F igure 3.12 . ). Yield strength of reinforcement ( ) Minimum cover (c) Embedment Length ( No. 8 bars and smaller 75 ksi < ( Embedment length) (Lap splice length) (Max. clear spacing between adjacent lap splice bars) Table 3. 7 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 526 . The properties of the noncontact lap splice in the connection area Table 3. 527 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 528 . The properties of the noncontact lap splice in the connection area Table 3. 529 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 530 . The properties of the noncontact lap splice in the connection area Table 3. 531 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 532 . The properties of the noncontact lap splice in the connection area Table 3. 533 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 534 . The properties of the noncontact lap splice in the connection area Table 3. 535 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 536 . The properties of the noncontact lap splice in the connection area Table 3. 537 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 538 . The properties of the noncontact lap splice in the connection area Table 3. 539 . Minimum cover and embedment length of reinforcing bars embedded in the UHPC by Graybeal (2014) Table 3. 8 . The properties of the noncontact lap splice in the connection area

PAGE 50

37 Figure 3.8. Plan view of the full simulated deck specimen Figure 3.9. Section A A of plan view in Figure 3.8.

PAGE 51

38 Figure 3.11. Section C C of plan view in Figure 3.8. Figure 3.10. Section B B of plan view in Figure 3.8.

PAGE 52

39 Calculation of the Development Length of #5 Bars The development length of #5 bars that were used to build the simulated deck was calculated in accordance with the American Concrete Institute ( ACI ) Section 25.4.2. In the next calculation, part (a) controlled with the development length is 2.37 ft. Calculation of development length for #5 rebar in this project: According to ACI Section 25.4.2: is the greater of (a) or (b) : (a) Using Table 25.4.2.2 in ACI Section 25.4.2 concrete relative to normal weight concrete of the same compressive strength

PAGE 53

40 (b) 12 in . Figure 3.12 . illustrates the shear key. This transverse joint view was suggested by CDOT. Even though CDOT reported that the minimum deck thickness is 8 in . (CDOT Bridge Design Manual, 2018), they wanted to test this specific transverse joint with the 6 in . deck thick ness to get the result of the worst case scenarios. Also, as Lafarge company is a funder for this experiment, they have changed a simple thing in this transverse joint. Lafarge wanted to study this shape, but without any space in the bottom between the two adjacent decks. Therefore, the labor er s will not need to create a formwork in the bottom for casting the UHPC material. Figure 3.12. Transverse Joint (connection).

PAGE 54

41 Simulated Deck Formwork Preparation A wood form was built f or casting the normal concrete panels . T here were no negative issues for all four sides except the side for a shear key (connection area side). In order to finish the shear key side, 16 wood pieces , having a 1. 5 in. thickness for each, fastened together to complete a 2 ft . width ( s ee Figure 3.13 a . ). All four sides of the normal concrete form were covered by transparent tape from inside to prevent any bond between the fresh concrete and the wood forms ( s ee Figure 3.13 b . ). Also, the white oil based paint was used instead of transparent tape for the ground part of the form. Figure 3.14 . shows the completed wood form for castin g normal concrete panels. (a) Figure 3.13. Formwork preparation for casting normal concrete : (a) Shear key side of the formwork ; (b) Wood parts were being covered by transparent tape ( b )

PAGE 55

42 The formwork for casting the UHPC material in the connection area was easy due to the connected normal concrete panel s from the bottom. Figure 3 .15 . shows the completed formwork for casting UHPC materials. Two wood parts were used on each side, and they are held by several steel clamps. Also, the wood parts were covered with transparent tape. The plastic sheet was under the connection area to make sur e that there will not be any bond between the fresh UHPC material and the ground due to the small spaces between the two normal concrete panel s from the bottom. Figure 3 .16 . shows the steel rebar measurement s that are spacing at 3 in. center to center. Figure 3.14. Completed wood formwork for casting one normal concrete panel

PAGE 56

43 Figure 3.15. Completed wood formwork for casting UHPC material in connection area Figure 3.16. Reinforcement measurements before casting the UHPC material (3 in. center to center)

PAGE 57

44 Curing Processes For the normal concrete, the two panels cured directly by covering them with a concrete blanket after casting the concrete ( s ee Figure 3 .17 . ). Then, the rewetting of the concrete was ended by cold water for 14 days. After the 14 days, the panels were rewetted each 2 to 3 days , approximately , for 28 days. Figure 3.17. Curing the normal concrete panels by concrete blanket

PAGE 58

45 For the UHPC curing, after casting the UHPC material in the joint area, the simulated deck joints cured by covering them with multiple layers of plastic sheets to prevent surface dehydration ( s ee Figure 3 .18 . ). The cold water was used to rewet the UHPC joint every day for 7 days. A fter this , the UHPC joints were rewetted each 2 to 3 days , approximately , until the age of 28 days. It wa s apparent that the water existed under the plastic sheets each time before the next wetting. Finally, the UHPC joints were exposed to the ambient environment. Figure 3.18. Curing the UHPC joints by multiple plastic sheets

PAGE 59

46 Installing the I nstrumentation The 55 kips MTS machine was used to complete the test of all three full simulated deck specimen. This machine generated the point load which represents the wheel truck load. The load was monotonic loading until the specimen failed, and this load was controlled to influence the specimen by the displacement control rate of 0.0 393 in /min. Figure 3.19. and Figure 3.20. show the instrumentation plan in order to study and anal yze the behavior of the UHPC material in the connection area. Six strain gauges were used for one full simulated deck, and all of them were placed on the UHPC compression face. Figure 3.20. shows each of the three strain gauges in both sides next to the lo ad point. The distance between the center of the load point and the first strain gauge was 3.5 in. This distance made sure that all strain gauges would be placed in a location not above the reinforcing bars . In addition, the full simulated deck was instrumented with a PI gauge to capture the curvature in the tension reinforcement layer. Also, a vertical linear variable differential transformer (LVDT) was used to measure the displacement changes at the midspan of the full simulated deck. Because all devices above were connected with one computer ( see Figure 3.21 ), the need of the load cell was mandatory to ensure the reading of the load matches with other outcomes of these devices.

PAGE 60

47 Figure 3.20. Plan view of the simulated deck test setup Figure 3.19. Longitudinal section view of the simulated deck test setup

PAGE 61

48 Figure 3.21. Computer is being connected for the simulated deck test

PAGE 62

49 CHAPTER IV EXPERIMENTAL RESULTS This chapter reveal s the results of this experimental study. These results are summarized into two sections: materials test results and simulated deck results. Materials Test Results Two types of experimental tests were completed for b oth materials ( UHPC and normal concrete ) . These tests are compression cylinders tests and flexural beams tests. In addition, the UHPC material was tested for tensile strength by split tension cylinders tests. Compression Cylinders Tests As was explained i n C hapter III, two machines were used to achieve the compression cylinders tests. For both machines, 220K MTS and Forney , the results were obtained as force values in terms of l bf units, so the following equation was used to convert the force to the streng th in terms of psi units. Where: : The compressive strength (psi) : The ultimate applied force (Ibf) : The contact area of the cylinders (

PAGE 63

50 UHPC Compression Cylinders Tests Table 4.1. lists the results of the compression cylinders for the UHPC materials at 7 days. These results were obtained from the 220K MTS machine. Figure s 4.1 . to 4.3. show the failure shape after testing. The average ultimate load and compressive strength were 185.9 kip s and 14 . 8 ksi , respectively. The applied load was controlled by a displacement rate of 0.0393 in /min in the 220k MTS machine. Table 4.2. and Table 4.3. list the results of the UHPC compression cylinders at 14 and 28 days , respectively. The average compressive strength of the UHPC material used in this study is 26.6 ksi. Figure s 4.4 . to 4.6 . the failure shape of each UHPC compression cylinder that was tested at 14 days, and Figure s 4.7 . to 4.12 . show the UHPC compression cylinders that were tested at 28 days. The rate of the ap plied load was 440 l bs/sec 88 l bs/sec. Normal Concrete Compression Cylinders Tests The average compressive strength of the normal concrete is 4 .7 ksi . This value is obtained from testing six cylinders at 28 days. Also, the normal concrete material was te sted at 7 days, and the average compressive strength was 2 . 1 ksi , approximately. For more details about the normal concrete material test results, review Appendix B.

PAGE 64

51 Specimens ID Ultimate Load ( kip ) Compression Stress ( ksi ) 1 C UHPC 7d 184.8 14.7 2 C UHPC 7d 197.2 15.7 3 C UHPC 7d 175.6 14.0 Average 185.9 14.8 Specimens ID Ultimate Load ( kip ) Compression Stress ( ksi ) 1 C UHPC 14d 296.3 23.6 2 C UHPC 14d 305.3 24.3 3 C UHPC 14d 284.6 22.7 Average 295.4 23.5 Table 4. 1 . Results of UHPC compression cylinder test at 7 days Table 4. 2 . Results of UHPC compression cylinder test at 14 days

PAGE 65

52 Specimens ID Ultimate Load ( kip ) Compression Stress ( ksi ) 1 C UHPC 28d 341.34 27.2 2 C UHPC 28d 310.54 24.7 3 C UHPC 28d 320.87 25.5 4 C UHPC 28d 319.21 25.4 5 C UHPC 28d 361.65 28.8 6 C UHPC 28d 350.69 27.9 Average 334.05 26.6 Table 4. 3 . Results of UHPC compression cylinder test at 28 days

PAGE 66

53 Figure 4.1. The f irst UHPC compression cylinder at 7 days after testing (1 C UHPC 7d) Figure 4.2. The s econd UHPC compression cylinder at 7 days after testing (2 C UHPC 7d) Figure 4.3. The t hird UHPC compression cylinder at 7 days after testing (3 C UHPC 7d)

PAGE 67

54 Figure 4.4. The first UHPC compression cylinder at 14 days after testing (1 C UHPC 14d) : (a) Failure shape ; Figure 4.5. The second UHPC compression cylinder at 14 days after testing (2 C UHPC 14d) : (a) Failure shape ; Figure 4.6. The third UHPC compression cylinder at 14 days after testing (3 C UHPC 14 d ): (a) ( b ) ( b ) ( b ) (a) (a)

PAGE 68

55 Figure 4.7. The first UHPC compression cylinder at 28 days after testing (1 C UHPC 28d) : (a) Failure shape ; Figure 4.8. The second UHPC compression cylinder at 28 days after testing (2 C UHPC 28d) : (a) Failure shape ; Figure 4.9. The third UHPC compression cylinder at 28 days after testing (3 C UHPC 28d) : (a) Failure shape ; (a) ( b ) ( b ) ( b ) (a) (a)

PAGE 69

56 Figure 4.10. The fourth UHPC compression cylinder at 28 days after testing (4 C UHPC 28d) : (a) Failure shape ; Figure 4.11. The fifth UHPC compression cylinder at 28 days after testing (5 C UHPC 28d) : (a) Failure shape ; Figure 4.12. The sixth UHPC compression cylinder at 28 days after testing (6 C UHPC 28d) : (a) Failure shape ; ( b ) ( b ) (a) (a) ( b ) ( a )

PAGE 70

57 Flexural Beams Tests The 20k MTS machine was used to complete the flexural strength for both UHPC and conventional concrete materials. This machine provides the ultimate load value, and the value can be used to calculate the flexural strength as follows: Where: : modulus of r u pture (psi) : The ultimate applied force ( l bf) : Span length (in) : Beam width (in) : Beam depth (in) The applied load was controlled by the displacement rate of 0.0393 in /min in the 20k MTS machine. In fact, the above equation is derived from ASTM, but it was modified in this study with the notch in the middle of the flexural beam span for both materials tested, the UHPC and the conventional concrete. UHPC Flexural Beams Tests Table 4.4. lists the flexu ral strength of six UHPC prisms; all of them were tested at 73 days. The average flexural strength was 918.8 psi . Figure 4.13. shows the failure shape of each beam.

PAGE 71

58 Conventional Concrete Flexural Beams Tests Appendix B. includes the results of the flexural beam tests for the normal concrete. The average flexural strength was 234.4 psi. Specimens ID Ultimate Load ( l bf) Flexural Strength (psi) 1 B UHPC 1806.2 790.2 2 B UHPC 2340.3 1023.9 3 B UHPC 2447.0 1070.6 4 B UHPC 1602.9 701.3 5 B UHPC 1985.9 868.8 6 B UHPC 2418.0 1057.9 Average 2100.0 918.8 Table 4. 4 . Results of UHPC flexural beams

PAGE 72

59 (a) 1 B UHPC (b) 2 B UHPC (c) 3 B UHPC (d) 4 B UHPC (e) 5 B UHPC (f) 6 B UHPC Figure 4.13. The failure shape of each UHPC flexural beam

PAGE 73

60 Split tension Cylinders Tests The 220k MTS machine was used to complete the splitting tensile strength of the UHPC cylinders. This machine provides the ultimate load value, and the value can be used to calculate the splitting tensile strength as follows: Where: : splitting tensile strength (psi) : The ultimate applied force ( l bf) : length (in) : diameter (in) Table 4.5. lists the results of the split tension cylinders tests of three UHPC cylinders. The average splitting tensile strength is 2161.1 psi. Figure 4.14. shows the failure shapes. Specimens ID Ultimate Load ( l bf) Splitting Tensile Strength (psi) 1 C ST 108120.3 2151.0 2 C ST 96113.3 1912.1 3 C ST 121658.2 2420.3 Average 108630.6 2161.1 Table 4. 5 . Results of split tension cylinders t ests .

PAGE 74

61 Figure 4.14. The failure shape of each UHPC split tension cylinder (a) 1 C ST (c) 2 C ST (b) 3 C ST

PAGE 75

62 Simulated Deck Test Results The three fu ll simulated decks were tested under the same conditions. The 55k MTS machine generated monotonic loading which represents the wheel truck load in the real structure. Table 4.6. lists the ultimate load of each simulated deck test. From the load values p lus strain gauges, vertical LVDT, and PI gauge, it was possible to analyze more results about each simulated deck. In fact, the bending capacity values of the UHPC joint is the most important data in the simulated deck test in this experimental study because these values can achieve the object of this thesis. The maximum bending moment f or specimens SD 1, SD 2, and SD 3 were 292.42, 256.93, and 304.99 kip in , respectively ( s ee Table 4.6.). Figure s 4.15. to 4.17. show the relationship between the load values a nd the displacement for SD 1, SD 2, and SD 3 , respectively. All simulated decks show similar behavior of this relationship. The average displacement value under the peak load of three simulated decks was 0.46 in. Also, Figure s 4.18. to 4.20. display the load and strain diagrams for SD 1, SD 2, and SD 3 , respectively. In Figure 4.18. and Figure 4.20 . , some of the strain gauges were out of service due to one of two reasons. The first reason is the failure bond between the strain gauge and the surface of the UHPC material; t he second reason is that the steel fiber influenced the strain gauges, so the reading of the strain gauge was not reasonable. Furthermore, Figure s 4.21. to 4.23. illustrate the relationship between the mo ment values and the values of the PI gauge for SD 1, SD 2, and SD 3 , respectively.

PAGE 76

63 The failure mode of all simulated deck was conventional concrete crushing. The conventional concrete crushing began first with the panel that has two layers of three reinfor cing bars , then the crushing appeared in the other panel . Figure s 4.24. to 4.26. show the failure mode for SD 1, SD 2, and SD 3 , respectively. Therefore, the failure mode was not the bond failure between the UHPC material and conventional concrete. In fact , there were no visible cracks in the UHPC joint for all simulated deck tests. As the failure mode within the UHPC joint is important after the full simulated deck reached the ultimate load, this will be discussed in the next chapter. Specimens ID Ultimate Load (kip) (in) Moment ( k ip in) SD 1 17.72 0.354 292.42 SD 2 15.57 0.539 256.93 SD 3 18.48 0.480 304.99 Average 17.26 0.46 284.78 Table 4. 6 . Results of three full simulated decks tests .

PAGE 77

64 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Load (Kip) Mid Span Displacment (in) Figure 4.15. Load vs . Mid Span displacement diagram for SD 1 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 0.00 0.50 1.00 1.50 2.00 2.50 Load (Kip) Mid Span Displacment (in) Figure 4.16. Load vs . Mid Span displacement diagram for SD 2

PAGE 78

65 Figure 4.17. Load vs . Mid Span displacement diagram for SD 3 Figure 4.18. Load vs . Strain diagram for SD 1 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 0.00 0.02 0.04 0.06 0.08 0.10 Load (Kip) Strain (%) Strain Gauge 1 Strain Gauge 2 Strain Gauge 3 Strain Gauge 4 Strain Gauge 5 Strain Gauge 6 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 Load (Kip) Mid Span Displacment (in)

PAGE 79

66 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 0.00 0.05 0.10 0.15 0.20 Load (Kip) Strain (%) Strain Gauge 1 Strain Gauge 2 Strain Gauge 3 Strain Gauge 4 Strain Gauge 5 Strain Gauge 6 Figure 4.19. Load vs . Strain diagram for SD 2 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 0.00 0.02 0.04 0.06 0.08 Load (Kip) Strain (%) Strain Gauge 1 Strain Gauge 2 Strain Gauge 3 Strain Gauge 4 Strain Gauge 5 Strain Gauge 6 Figure 4.20. Load vs . Strain diagram for SD 3

PAGE 80

67 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Moment (Kip in) PI gauge (%) Figure 4.21. Moment vs . PI gauge for SD 1 0.00 50.00 100.00 150.00 200.00 250.00 300.00 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 Moment (Kip in) PI gauge (%) Figure 4.22. Moment vs . PI gauge for SD 2

PAGE 81

68 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 0.000 0.002 0.004 0.006 0.008 0.010 0.012 0.014 0.016 0.018 0.020 Moment (Kip in) PI gauge Figure 4.23. Moment vs . PI gauge for SD 3 Figure 4.24. The failure mode of the normal concrete being crushed in SD 1 Conventional Concrete Crushing Panel with 3 rebars /layer

PAGE 82

69 Figure 4.25. The failure mode of the normal concrete being crushed in SD 2 Figure 4.26. The failure mode of the normal concrete being crushed in SD 3 Panel with 3 rebars /layer Panel with 3 rebars /layer Conventional Concrete Crushing Conventional Concrete Crushing

PAGE 83

70 CHAPTER V ANALYSIS AND DISCUSSION OF RESULTS Chapter V includes the analysis and discussion of test results of this experimental program, and this will be into two parts: material strength and simulated deck strength. Material Strength Compressive Strength The average compressive strength for UHPC material in this study was 14.8 ksi at 7 days, 23.5 ksi at 14 days, and 26.6 ksi at 28 days. In fact, the compressiv e strength for the UHPC materials should be more than 20 ksi at 28 days in order to say that this material is in the UHPC range. This is based in what the U.S. Federal Highway Administration (FHWA) reported. Table 5.1. lists the UHPC material properties ra nge in FHWA ( Russell and Graybeal, 2013). Property Range Compressive strength 20 to 30 ksi Tensile cracking strength 0.9 to 1.5 ksi Modulus of elasticity 6000 to 10000 ksi 0.2 Coefficient of thermal expansion 5.5 to 8.5 millionths/ o F Creep coefficient 0.2 to 0.8 Total shrinkage Up to 900 millionths Table 5. 1 . The range of the UHPC material properties by FHWA. ( Russell and Graybeal, 2013)

PAGE 84

71 Therefore, the average compressive strength of UHPC at 28 days in this study achieved the required value in the F HWA . Furthermore, referring to Table 3.3, Lafarge, which is the UHPC material provider for this study, reported that Ductal UHPC should be more than 21 ksi at 28 days, and 14 ksi at four days. In fact, the average compressive strength at 14 days is 23.5 ksi , which is 88.3 percent of the strength at 28 days (26.6 ksi). This percentage may lead to changes in any bridge project that used a precast deck ed bulb tee bridge girder system because the project may switch to the next construction process after 14 days instead of 28 days of UHPC casting. In addition , the compressive strength of the UHPC material is more than the conventi onal concrete in this experimental program by approximately 82 percent at 28 days, and 85 percent at 7 days. Modulus of Elasticity As the compressive strength of UHPC material is known, the modulus of elasticity can be calculated. According to F HWA ( Russ ell and Graybeal, 2013) , the modulus of elasticity for the UHPC material that has a compressive strength between 4 to 28 ksi can be calculated based on the next equation: in psi units Where: : modulus of elasticity : The UHPC compressive strength

PAGE 85

72 Consequently, the modulus of elasticity of the UHPC material in this study was 7534. 9 ksi at 28 days, based on the previous equation. Also, referring to table 5.1., this value of modulus of ela sticity is in the range of the UHPC material properties by FHWA (6000 to 10000 ksi). In addition, Table 3.3 (Lafarge, 2018) shows that the modulus of elasticity of the Ductal UHPC material that was used in this study should be more than 6500 ksi at 28 day s. Simulated Deck Strength This experimental study investigated the use of the UHPC material in the connection joint between the precast deck ed bulb tee bridge girders. Therefore, the results of a conventional concrete deck (Cast in place deck) is require d to create a comparison with the results of this experimental study. Consequently, Appendix A includes the results of the design and analysis of nominal moment strength of a conventional concrete deck (Cast in place deck) that had similar details of the s imulated bulb tee bridge girder deck which was studied experimentally in this thesis. The results of Appendix A show that the nominal moment strength of a conventional concrete deck (Cast in place deck) is 16.42 kip ft (197.04 kip in) that gives ultimate load is 11.94 kips. Therefore, the deck of the bulb tee bridge girder that has UHPC material for the connection joints is stronger than the cast in place deck (conventional concrete deck) by approximately 30 percent in terms of the bending moment capacity .

PAGE 86

73 Figure 5.1. illustrates the relationship between the bending moment and the displacement of each simulated deck. The blue line in F igure 5.1. represents the average bending moments and displacements of all simulated decks, and the black dash li ne shows the level of the calculated nominal bending moment for the cast in place deck that has similar specifications as the simulated bulb tee girder deck in this experimental study. Figure 5. 1 . The bending moment vs. displacemen t for all simulated decks Figure 5. 1 . Destroyed shape inside the UHPC joint of SD 1. Figure 5. 2 . The bending moment vs displacement for all simulated decks. 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Moment (Kip in) Mid Span Displacment (in) SD-1 SD-2 SD-3 Average Calculated capacity for cast-in-place deck

PAGE 87

74 As was mentioned in C hapter IV, the failure mode of all simulated decks was the conventional concrete crushing. However, it was important to know the actions into the UHPC connection after the simulated deck reached its ultimate load and crushing of the co nventional concrete. Consequently, the monotonic load was applied on the simulated deck and regulated by the displacement rate of 0.0393 in /min until the simulated deck reached the peak load. After this, the load was continuously applied with different di splacement rates in order to check the reinforcement response inside the UHPC connection. In fact, the reinforc ing bars were yielded for all simulated decks , and it has appeared inside the UHPC joint in each simulated deck that there were cracks around the reinforc ing bars in the tension layer of the concrete panel that has three re inforcing bars ( s ee Figure s 5.2 . to 5.4 . ). At this point, the hydraulic ram reached the limit, so it was not abl e to complete the demolition of t he simulated deck. In F igure 5.4., one steel rebar in the SD 1 broke the UHPC material, and this rebar was located under the concentrated load directly. Expectedly , the bond strength between the rebar and the UHPC material is good enough where there was n o pullout failure in the connection area. In addition, there were no visual cracks in the external shape of the UHPC block.

PAGE 88

75 Figure 5. 2 . Destroyed shape inside the UHPC joint of SD 1 Figure 5. 3 . Destroyed shape inside the UHPC joint of SD 2. Figure 5. 4 . Destroyed shape inside the UHPC joint of SD 1. Figure 5. 3 . Destroyed shape inside the UHPC joint of SD 2

PAGE 89

76 Figure 5. 4 . Destroyed shape inside the UHPC joint of SD 3

PAGE 90

77 CHAPTER VI CONCLUSIONS, RECOMMENDATIONS AND FUTURE RESEARCH Conclusions Goal The goal of this project was to investigate the behavior of the UHPC material for the connection of precast decked bulb tee bridge girders using noncontact lap splice and specific transverse joint detail (shear key). The designing and building of custom si mulated deck specimens was necessary to complete this investigation. Material s Finding the two different concrete material types for this experimental program was the first step. Th e Ductal UHPC was chosen to represent the ultra high performance concrete , and the 4000 psi Quikrete w as used to represent the conventional concrete. In addition, the reinforc ing bars that were used in this project are epoxy coated #5. A combination of standard material strength tests were conducted to determine the compressiv e strength and flexural strength of the UHPC material and conventional concrete. The average compressive strength of the UHPC material and the conventional concrete were 26.6 ksi and 4.69 ksi , respectively. This means that the UHPC material had about six times more compressive strength when compared to the conventional concrete in this study . Simulated bulb tee bridge girder deck Three full simulated deck specimens were tested to investigate the UHPC material in the connection area between precast dec ked bulb tee bridge girders. Each simulated deck had two

PAGE 91

78 panels of conventional concrete, and each panel represents a precast deck of the bulb tee girder. Each panel was longitudinally reinforced with two layers of the #5 epoxy coated rebar. The first pane l had three bars in each layer, and the second panel had four bars in each layer. The design of the simulated deck specimen was based on three factors. The first factor is the noncontact lap splice. For this factor, the design was according to the guidance published by the U.S. Federal Highway Administration (FHWA). The second factor is the development length of the #5 epoxy coated rebar. The third factor is the transverse joint detail that was proposed by CDOT. All simulated deck specimens were tested unde r three bending moments with monotonic load, and they show ed great strength compared to the cast in place deck (no connection) that is designed and analyzed theoretically in this project. In fact, it was observed that the strength of the deck of the bulb tee girder with UHPC connection is stronger than the cast in place deck (conventional co ncrete deck) by approximately 30 percent i n terms of bending moment capacity . In addition, the failure mode of all simulated deck specimen s was the result of the conventional concrete crushing . V isually , there were no cracks of the external shape of UHPC m aterial in the connection. Furthermore, the UHPC material showed enough bond to the reinforcing bars. Recommendations UHPC Preparation The UHPC mix operation is one of the most important steps to work with this kind of material. UHPC mixing takes more time than mixing normal concrete due to the low water cement ratio. Therefore, this mixing needs a good mixer such as the pan mixer used in this

PAGE 92

79 experimental program. Critically , great care must be taken to ensure that no balls of steel fibers are formed b y pouring the steel fiber too quickly during the UHPC mixing process. Structural Design The particular transverse joint detail (shear key) for the connection area between the deck s of bulb tee girders in this study was able to show enough bond strength be tween the faces of the two different materials -UHPC and conventional concrete -until the failure mode concrete crushing was observed. T he noncontact lap splice is recommended to be design ed according to the guidance published by the U.S. Federal H ighway A dministration (FHWA) . This design will provide satisfactory performance for connecting decked bulb tee girders by using UHPC material. Construction Considerations In fact, one of the primary goals of decked bulb tee bridge girder s is to decrease the construction time on the site . This can be achieved by eliminating the processes of the cast in place deck. On the other hand, the bulb tee bridge girder still needs to have a connection that is c ast in place material, and this operation needs formwork. But the design of the transverse joint of this study also will eliminate the formwork process for casting the connection material on the sit e; t herefore, the construction time can be reduced even mo re. Future Research The experimental program of this thesis only focused on the bending capacity of using UHPC in the connection area between bridge decks. Consequently, the investigation of the shear capacity between the two materials is very important.

PAGE 93

80 Th is experimental study only focused on one dimension of the joint width , which is 6 in. Therefore, it is significant to know the effect of different width s of the joint area on the bending capacity. The curing of the UHPC material in the connection area in this experimental program was for approximately 28 days. Meanwhile, the compressive strength of the UHPC material in this study was figured at 28 and 14 days, and it was detected that the compressiv e strength at 14 days was approximately 88 percent of the compressive strength at 28 days. Based on this, it would be valuable to investigate the UHPC material in the connection area at 14 days instead of 28 days, because this will help to reduce the const ruction time.

PAGE 94

81 REFERENCES ACI. (2014). Building Code Requirements for Structural Concrete (ACI 318 14) . Farmington Hills, MI. ASTM C192 / C192M , West Conshohocken, PA. ASTM C39 / C39M , West Conshohocken, PA. ASTM C496 / C496M , West Conshohocken, PA. third ASTM C78 / C78M , West Conshohocken, PA. Colorado Department of Transportation. (2018). Bridge Desig n Manual , Denver, CO. Grace, N. F., Bebawy, M., and Kasabasic, M. (2015). Evaluation and Analysis of Decked Bulb T Beam Bridges (No. RC 1620). Department of Civil Engineering, Lawrence Technological University. Graybeal, B. A. (2006). Material property characterization of ultra high performance concrete (No. FHWA HRT 06 103). Graybeal, B.A. (2014). Design and construction of field cast UHPC connections (No. FHWA HRT 14 084). Haber, Z. B., and Graybeal, B. A. (201 8). Lap Spliced Rebar Connections with UHPC Closures. Journal of Bridge Engineering , 23 (6), 04018028. LafargeHolcim. Ductal JS1000 (ver 2.1). [Brochure]. Chicago, I L . Prem, P. R., Ramachandra Murthy, A., and Bharatkumar , B. H. (2015). Influence of curing regime and steel fibers on the mechanical properties of UHPC. Magazine of Concrete Research , 67 (18), 988 1002. Russell, H. G., and Graybeal, B. A. (2013). Ultra high performance concrete: A state of the art report for th e bridge community (No. FHWA HRT 13 060).

PAGE 95

82 Yuan, J., and Graybeal, B. A. (2014). Bond behavior of reinforcing steel in ultra high performance concrete (No. FHWA HRT 14 090).

PAGE 96

83 Appendix A. Design and Analysis of Cast in Place Deck (No Connection) Overview This a ppendix reported the design and analysis of nominal moment strength of a deck that is continuous ( c ast in place deck), so there is no connection (joint area). As a result of details of the simulated bulb tee bridge girder deck which was studied experimentally in this thesis, the section in Appendix Figure A.1. was designed. Therefore , the design and analysis of this appendix accomplished the comparison between u sing UHPC material to connect two adjacent precast deck ed bulb tee bridge girders and cast in place deck ( n o connection) . In order to complete the design and analysis of nominal moment strength of the section in Appendix Figure A.1., the calculation will be carried out by SAP2000 software and analysis of doubly reinforced beam sections. Appendix Table A.1. lists the properties of the section in Appendix Figure A.1. Appendix Figure A.1. A section that represents the cast in place deck ( n o connection)

PAGE 97

84 Variable Description Value Area of tension reinforcement. = Area of compression reinforcement. = Width of compression face of member. = Distance from extreme compression fiber to centroid of tension reinforcement. = Distance from extreme compression fiber to centroid of compression reinforcement. = Modulus of elasticity of reinforcement. = Compressive stress in concrete = Specified yield strength of nonprestressed reinforcement = Span length = Ratio of depth of rectangular stress block, to depth to neutral axis, . = Maximum useable compressive strain for concrete = Appendix Table A.1. The properties of the section in Appendix Figure A.1.

PAGE 98

85 Analysis of D oubly R einforced B eam S ections The following calculation is to find a nominal moment strength and ultimate load by analysis of a doubly reinforced beam section in Appendix Figure A.1. with the properties in Appendix Table A.1.: 1 . Assume that the steel in tension area is yielding Where: 2 . Select a value for Note: value is taken as after many iterative procedures between and 3 . Calculate the compress ion reinforcement strain ( 4 . Calculate the stre ss in compression reinforcement

PAGE 99

86 5 . Calculate the force in compression reinforcement ( 6 . Calculate concrete compression force ( ) 7 . Calculate force in tension reinforcement ( ) 8 . Check the equilibrium of the section 9 . Confirm that the tension steel is yielding 10 . Calculate the depth of the compression stress block ( )

PAGE 100

87 11 . Calculate the nominal moment strength ( ) Where : the strength reduction factor, for this section is 0.9 12 . Calculate the ultimate load ( For Simply support ed beam :

PAGE 101

88 SAP2000 S oftware for Cast in P lace D eck A nalysis This section is about the calculation of the moment capacity of the section in Appendix Figure A.1. with the properties in Appendix Table A.1. using the moment curvature method in the SAP2000 software. Appendix Figure A.2. shows the maximum bending moment that is 16.8 kip ft. This value is approximately similar to the result of the analysis of doubly reinforced beam sections (16.4 kip ft). Appendix Figure A. 3 . illustrates the materials properties entered in SAP2000. Appendix Figure A. 2 . T he moment curvature of the section in Appendix Figure A.1. with the properties in Appendix Table A .1 . using SAP2000

PAGE 102

89 Appendix Figure A. 3 . Material properties in the SAP2000 in order to analyze the section in Appendix Figure A.1.

PAGE 103

90 Appendix B . Additional Experimental Test Results Normal Concrete Material Tests Normal concrete compression cylinder tests Appendix T able B.1. lists the results of the compression cylinders for the normal concrete materials at 7 days, and Appendix F igure B.1. shows the failure shape after testing. For the normal concrete tests at 28 days, Appendix T able B.2. shows the results of six nor mal concrete cylinders. Appendix Figures B.2. to B.7. show the failure shape of each cylinder after testing at 28 days. The 220 K MTS machine was used to complete the compression cylinder test at 7 days, and the Forney machine was used for testing at 28 da ys. Specimens Load ( kip ) Compression Stress ( ksi ) 1 C NC 7d 23.2 1.8 2 C NC 7d 23.0 1.8 3 C NC 7d 34.3 2.7 Average 26.8 2.1 Appendix table B.1. Results of normal concrete compression cylinder test at 7 days

PAGE 104

91 Specimens Load ( kip ) Compression Stress ( ksi ) 1 C NC 28d 61.0 4.9 2 C NC 28d 50.1 4.0 3 C NC 28d 63.8 5.1 4 C NC 28d 51.2 4.1 5 C NC 28d 55.2 4.4 6 C NC 28d 72.4 5.8 Average 58.9 4.7 Appendix table B.2. Results of normal concrete compression cylinder test at 28 days

PAGE 105

92 (a) (b) (c) Appendix figure B.1. The failure shape of three normal concrete cylinders after testing at 7 days. (a) 1 C NC 7d , (b) 2 C NC 7d , (c) 3 C NC 7d

PAGE 106

93 Appendix figure B .2. The first normal concrete compression n cylinder at 28 days after testing ( 1 C NC 28d Appendix figure B . 3 . The second normal concrete compression cylinder at 28 days after testing ( 2 C NC 28d Appendix figure B .4. The third normal concrete compression cylinder at 28 days after testing ( 3 C NC 28d (a) (b) (b) (b) (a) (a)

PAGE 107

94 Appendix figure B . 5 . The fourth normal concrete compression cylinder at 28 days after testing ( 4 C NC 28d ): (a) Failure shape; (b) Forney Appendix figure B . 6 . The fifth normal concrete compression cylinder at 28 days after testing ( 5 C NC 28d Appendix figure B . 7 . The sixth normal concrete compression cylinder at 28 days after testing ( 6 C NC 28d (a) (b) (b) (b) (a) (a)

PAGE 108

95 Normal concrete flexural prisms tests Appendix Table B.3. lists the flexural strength of six conventional concrete prisms, all of which were tested at 80 days. The average flexural strength was 234.4 psi. Appendix F igure B .8. shows the failure shape of each beam. Specimens ID Ultimate Load ( l bf) Flexural Strength (psi) 1 B NC 514.8 225.23 2 B NC 558.38 244.29 3 B NC 534.1 233.67 Average 535.8 234.4 Appendix Table B .3. Results of normal concrete flexural beams

PAGE 109

96 (a) 1 B NC (b) 2 B NC (c) 3 B NC Appendix F igure B . 8. The failure shape of each normal concrete flexural prism