<%BANNER%>

Load Test and Distribution of Tire Loads in a Gfrp Composite Bridge Deck

Permanent Link: http://ufdc.ufl.edu/UFE0042632/00001

Material Information

Title: Load Test and Distribution of Tire Loads in a Gfrp Composite Bridge Deck
Physical Description: 1 online resource (107 p.)
Language: english
Creator: Singh, Abhay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bridge, composites, deck, fiber, gps, load, polymer, reinforced, repair, test
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Civil Engineering thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Florida has largest inventory of moveable bridges in the nation. Most of them employ steel grid as a bridge deck for part of their span. Worn steel grid decks have poor skid resistance, especially in rain, and provide poor riding comfort. Furthermore they have high maintenance cost and high noise levels when traffic travels across them. The Florida Department of Transportation (FDOT) is considering replacing worn steel grid deck with glass fiber reinforced polymer (GFRP) deck. GFRP deck systems provide a solid riding surface thereby addressing noise and stopping distance concerns with worn steel grid decks. Furthermore, weight and dimensions (thickness) of GFRP decks are comparable to the steel grid deck. Which allow it to be used as a direct replacement for steel grid decks. GFRP deck systems, however, are relatively new to the bridge infrastructure industry. As such, there is concern regarding the durability and field performance of GFRP bridge decks. To evaluate the durability and field performance of these new systems, FDOT replaced an existing steel grid deck system with glass fiber reinforced polymer (GFRP) deck in Belle Glade, FL in Aug. ? Sept. 2009. This thesis presents a detailed analysis of the bridge deck behavior based on the results of the first load test. Soon after opening the bridge for service, a load test was conducted on the bridge to determine the behavior of both the GFRP deck and underlying structural steel girders. GFRP bridge deck was instrumented prior to installation for both load test and long-term monitoring. The use of an interactive GPS system allowed the strain-truck position to be plotted and presented. One of the major findings was that wheel load response of the bridge was very local, this is an indicator of the new deck system is flexible compare to the conventional bridge decks (steel and concrete). Maximum strain recorded during the field test for the design wheel load (AASHTO LRFD) was 15% of the failure strain. From the measured load test data, it was observed that both (steel and GFRP) material remains linear ? elastic. Existing steel girders system showed a factor of safety of 4.4. Bridge was built in 1950?s, from the factor of safety observed for the existing superstructure, it can be concluded that steel girder system is still performing well. Contribution of the top plate to the flexural stiffness of the two part GFRP deck system was revealed through the composite action between each component. The strains measured during the field load test were found to match laboratory tests well. Strains were calibrated to the wheel load to be used for the purpose of converting the monitoring strain into the wheel loads. Performance of the existing steel girder system was evaluated. Results presented in this thesis and future research associated with the project will provide a better understanding of the behavior of GFRP deck system and their possible use as a replacement of deteriorated steel grid decks and other conventional decks.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Abhay Singh.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Hamilton, Homer R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042632:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042632/00001

Material Information

Title: Load Test and Distribution of Tire Loads in a Gfrp Composite Bridge Deck
Physical Description: 1 online resource (107 p.)
Language: english
Creator: Singh, Abhay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bridge, composites, deck, fiber, gps, load, polymer, reinforced, repair, test
Civil and Coastal Engineering -- Dissertations, Academic -- UF
Genre: Civil Engineering thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Florida has largest inventory of moveable bridges in the nation. Most of them employ steel grid as a bridge deck for part of their span. Worn steel grid decks have poor skid resistance, especially in rain, and provide poor riding comfort. Furthermore they have high maintenance cost and high noise levels when traffic travels across them. The Florida Department of Transportation (FDOT) is considering replacing worn steel grid deck with glass fiber reinforced polymer (GFRP) deck. GFRP deck systems provide a solid riding surface thereby addressing noise and stopping distance concerns with worn steel grid decks. Furthermore, weight and dimensions (thickness) of GFRP decks are comparable to the steel grid deck. Which allow it to be used as a direct replacement for steel grid decks. GFRP deck systems, however, are relatively new to the bridge infrastructure industry. As such, there is concern regarding the durability and field performance of GFRP bridge decks. To evaluate the durability and field performance of these new systems, FDOT replaced an existing steel grid deck system with glass fiber reinforced polymer (GFRP) deck in Belle Glade, FL in Aug. ? Sept. 2009. This thesis presents a detailed analysis of the bridge deck behavior based on the results of the first load test. Soon after opening the bridge for service, a load test was conducted on the bridge to determine the behavior of both the GFRP deck and underlying structural steel girders. GFRP bridge deck was instrumented prior to installation for both load test and long-term monitoring. The use of an interactive GPS system allowed the strain-truck position to be plotted and presented. One of the major findings was that wheel load response of the bridge was very local, this is an indicator of the new deck system is flexible compare to the conventional bridge decks (steel and concrete). Maximum strain recorded during the field test for the design wheel load (AASHTO LRFD) was 15% of the failure strain. From the measured load test data, it was observed that both (steel and GFRP) material remains linear ? elastic. Existing steel girders system showed a factor of safety of 4.4. Bridge was built in 1950?s, from the factor of safety observed for the existing superstructure, it can be concluded that steel girder system is still performing well. Contribution of the top plate to the flexural stiffness of the two part GFRP deck system was revealed through the composite action between each component. The strains measured during the field load test were found to match laboratory tests well. Strains were calibrated to the wheel load to be used for the purpose of converting the monitoring strain into the wheel loads. Performance of the existing steel girder system was evaluated. Results presented in this thesis and future research associated with the project will provide a better understanding of the behavior of GFRP deck system and their possible use as a replacement of deteriorated steel grid decks and other conventional decks.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Abhay Singh.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Hamilton, Homer R.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042632:00001


This item has the following downloads:


Full Text

PAGE 1

LOAD TES T AND DISTRIBUTION OF TIRE LOADS IN A GFRP COMPOSITE BRIDGE DECK By ABHAY PRATAP SINGH A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2010 1

PAGE 2

2010 Abhay Pratap S ingh 2

PAGE 3

To m y Mummy, Papa and Hans 3

PAGE 4

ACKNOWL EDGMENTS I would like to thank Dr. H.R. Hamilton for his motivation and guidance throughout my research. His continuous encouragement for inde pendent working and critical thinking led to countless interesting findings. I w ould also like to thank Florid a Department of Transportation (FDOT) for funding this research. I would also like to extend my thanks to the FDOT Structures Research Center staff includi ng Mr. Marcus Ansley (deceased), David Allen, Stephen Eudy, Anthony Hobbs, Kyle Ramsdell, Paul Tighe, Da vid Wagner, and Chris Weigly for their outstanding efforts in preparation and execution of the bridge load test reported in this thesis. I would also like to thank Dr. G.R. Consolaz io and Dr. David Prevatt for serving on my supervisory committee and for providing valuable suggestions and feedback throughout the research process. Finally, I would like to thank my family and friends for their continuous support and confidence. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................8LIST OF ABBREVIATIONS ........................................................................................................11ABSTRACT ...................................................................................................................................12 CHAP TER 1 INTRODUCTION................................................................................................................. .14Introduction .............................................................................................................................14Objectives ...............................................................................................................................16Scope of Work ........................................................................................................................162 EXISTING BRIDGE CONFIGURATION............................................................................173 GFRP DECK DESIGN...........................................................................................................22Deck Design ............................................................................................................................22Deck Installation .....................................................................................................................224 INSTRUMENTATION AND DATA ACQUISITION..........................................................33Approach .................................................................................................................................33Strain Gages ............................................................................................................................34Thermocouples .......................................................................................................................35Displacement Gages ...............................................................................................................36Instrumentation Positions .......................................................................................................37Sampling Rate .........................................................................................................................37Data Acquisition System ........................................................................................................385 BRIDGE LOAD TEST OCTOBER, 2009..........................................................................51Objectives ...............................................................................................................................51Truck Positions and Load Levels ...........................................................................................51Test Setup ...............................................................................................................................52Test Procedure ........................................................................................................................53Step 1 Static Load Test .................................................................................................53Step 2 Rolling Load Test ..............................................................................................53 5

PAGE 6

6 BRIDGE TEST RESULTS OCTOBER, 2009.....................................................................59Overview .................................................................................................................................59Static Load Behavior ..............................................................................................................59Rolling Load Behavior ...........................................................................................................60Load Displacement ...............................................................................................................64Composite Behavior ...............................................................................................................66Performance of the Existing Steel Superstructure ..................................................................68Field Test versus Laboratory Test ..........................................................................................70Load Distribution ....................................................................................................................71Distribution Factors Ba sed on Flexural Strain ................................................................72Distribution Factors Based on Shear Strain .....................................................................73Comparison of Distribution Factors ................................................................................74Load Strain Calibration Curves ...........................................................................................757 SUMMARY AND CONCLUSION.....................................................................................102APPENDIX:DETAILED LOAD TEST PROCEDURE.............................................................104LIST OF REFERENCES .............................................................................................................106BIOGRAPHICAL SKETCH .......................................................................................................107 6

PAGE 7

LIST OF TABLES Table page 4-1 Summary of instrumentation for load tests and monitoring. .............................................404-2 Coordinate position of gages. ............................................................................................404-3 Sampling rate calculations. ................................................................................................415-1 Test truck axle loads (kip). .................................................................................................546-1 Maximum static strain values. ...........................................................................................776-2 Load distribution factors. ...................................................................................................77 7

PAGE 8

LIST OF FI GURES Figure page 2-1 Existing bridge. ..................................................................................................................192-2 Elevation view of main span ..............................................................................................202-3 Damaged and repaired existing steel grid deck .................................................................202-4 Existing framing layout for lift out span. ...........................................................................212-5 Cross sectional view of lift out span. .................................................................................213-1 GFRP deck configuration ..................................................................................................253-2 Existing steel grid deck ......................................................................................................253-3 GFRP bottom section panel layout. ...................................................................................263-4 Layout of GFRP panels ......................................................................................................273-5 Top GFRP section layout. ..................................................................................................283-6 Formwork for grout pads ...................................................................................................293-7 Installation of bottom GFRP panels ...................................................................................293-8 Transition between GFRP deck and concrete deck ...........................................................303-9 Grout pockets being poured at each stud ...........................................................................303-10 Median anchors ..................................................................................................................313-11 Top GFRP panels ...............................................................................................................313-12 Placement of polymer concrete wearing surface ...............................................................313-13 Bridge in service after construction ...................................................................................324-1 Two northbound lanes showing truck traffic marks on the road surface ...........................424-2 Origin for the GPS .............................................................................................................434-3 FBS gages mounted on steel girders. .................................................................................444-4 FBS gage ............................................................................................................................444-5 Installed FBS gage on the steel girder ...............................................................................45 8

PAGE 9

4-6 Location of instrumented panels (B9 and B10) thermocouples and displacement gages. .................................................................................................................................454-7 Position of bonded strain gages and rosettes on GFRP deck panels B9 and B10. ............464-8 Instrumentation on the GFRP Deck ...................................................................................464-9 Thermocouples on the surface of the GFRP deck .............................................................474-10 Displacement gages on the steel girders ............................................................................484-11 LVDT on the GFRP deck. .................................................................................................484-12 Monitoring DAQ (cRIO) and various input modules ........................................................494-13 Location of the cRIO and instrumentation wiring .............................................................505-1 FDOT utility truck used for Field Load Test .....................................................................555-2 Truck in position TP1. .......................................................................................................555-3 Truck in position TP2. .......................................................................................................555-4 Truck in position TP3. .......................................................................................................565-5 Truck in position TP4. .......................................................................................................565-6 Truck in position TP5. .......................................................................................................565-7 Truck positions for bridge test. ..........................................................................................575-8 Truck position referenc e marks on the bridge deck ...........................................................575-9 Location of the GPS dome and rela tive location of the truck axles. ..................................585-10 Rolling Test Flow chart. .................................................................................................586-1 Typical strain-time history. ................................................................................................786-2 Influence lines for positive bending ...................................................................................796-3 Influence lines for negative bending ..................................................................................806-4 Distance between axles ......................................................................................................816-5 Influence lines for rosette ...................................................................................................826-6 Influence lines for gage S5 and S6 for TP1. ......................................................................836-7 Relative location of gages S5 and S6 and maximum strain. ..............................................83 9

PAGE 10

6-8 Influence lines for gage S4and S7 for TP5. .......................................................................846-9 Relative location of gages S4 and S7 and maximum strain. ..............................................846-10 Partial influence lines for soffit gage S7 ............................................................................856-11 Load displacement for the bridge deck. ..........................................................................856-12 Displacement time history for the steel girders. ................................................................866-13 GFRP deck Modulus map. .................................................................................................866-14 N.A. locations for panel B10 .............................................................................................876-15 N. A. locations for panel B9 ..............................................................................................886-16 Location of measured and calculated elastic N.A. .............................................................896-17 Strain in extreme bottom fiber of steel girder. ...................................................................906-18 Strain influence lin es for full bridge gage ..........................................................................916-19 Test set up ..........................................................................................................................926-20 Instrumentation and appli cation of load for lab test ..........................................................926-21 Lab test ...............................................................................................................................936-22 Typical influence line illustrating calculation of di stribution factor. ................................946-23 Modified influence lines for distribution factor calculations .............................................956-24 Partial influence lines fo r web gage R1 at axle P5. ...........................................................966-25 Strain rosettes reflecting ....................................................................................................976-26 Comparison of Influence lines for soffit and web gage for 18 kip of wheel load. ............986-27 Modified influence lines for distribution factor calculations .............................................996-28 Load Strain Calibration Curve for positive bending gage. ..............................................1006-29 Load Strain Calibration Curve for negative bending gage ..............................................1006-30 Load Stress plots for full bridge gages .........................................................................101 10

PAGE 11

LIST OF ABBRE VIATIONS a Measured strain in 0 degree strain gages of the 0 45 90 degree rosette b Measured strain in 45 de gree strain gages of the 0 45 90 degree rosette c Measured strain in 90 de gree strain gages of the 0 45 90 degree rosette xy Calculated shear strain 11

PAGE 12

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering LOAD TEST AND DISTRIBUTION OF TIRE LOADS IN A GFRP COMPOSITE BRIDGE DECK By Abhay Pratap Singh December 2010 Chair: H.R. (Trey) Hamilton Major: Civil Engineering Florida has largest inventory of moveable bridges in the nation. Most of them employ steel grid as a bridge deck for pa rt of their span. Worn steel grid decks have poor skid resistance, especially in rain, and provide poor riding comfort. Furthermore they have high maintenance cost and high noise levels when traffic trav els across them. The Florida Department of Transportation (FDOT) is considerin g replacing worn steel grid deck with glass fiber reinforced polymer (GFRP) deck. GFRP deck systems pr ovide a solid riding surf ace thereby addressing noise and stopping distance concerns with worn steel grid decks. Furthermore, weight and dimensions (thickness) of GFRP decks are comparable to the stee l grid deck. Which allow it to be used as a direct replacement for steel grid de cks. GFRP deck systems, however, are relatively new to the bridge infrastructure industry. As such, there is con cern regarding the durability and field performance of GFRP bridge decks. To evaluate the durability and field performance of these new systems, FDOT replaced an existing steel grid deck system with glass fiber reinforced polymer (GFRP) deck in Belle Glade, FL in Aug. Sept. 2009. This thesis presents a de tailed analysis of the bridge deck behavior based on the results of the first load test. Soon after opening the bridge for service, a load test was conducted on the bridge to determine the behavior of both the GFRP deck and underlying 12

PAGE 13

13 structural steel girders. GFRP bridge deck was instrumented pr ior to installation for both load test and long-term monitoring. The use of an in teractive GPS system allowed the strain-truck position to be plotted and presented. One of the major findings was that wheel load response of the bridge was very local, this is an indicator of the new deck system is flexible compare to the conventional bridge decks (steel and concrete). Maximum strain recorded during the field test for the design wheel load (AASHTO LRFD) was 15% of the failure strain. From the measured load test data, it was observed th at both (steel and GFRP) material remains linear elastic. Existing steel girders system showed a factor of safety of 4.4. Bridge was built in 1950s, from the factor of safety observed for the existing supe rstructure, it can be concluded that steel girder system is still performing well. Contribution of the top plate to the flexural stiffness of the two part GFRP deck system was revealed through the composite action between each component. The strains measured during the fi eld load test were found to matc h laboratory tests well. Strains were calibrated to the wheel load to be used for the purpose of converting the monitoring strain into the wheel loads. Performance of the existing steel girder system was evaluated. Results presented in this thesis and future research a ssociated with the projec t will provide a better understanding of the behavior of GFRP deck syst em and their possible use as a replacement of deteriorated steel grid decks and other conventional decks.

PAGE 14

CHAP TER 1 INTRODUCTION Introduction Florida has the largest inventor y of moveable bridges in the nation, with a total of 148, of which 91% are bascule, 7% are swing and 2% are lift bridges; Most employ open grid steel decks as a riding surface for part of their span (National Bridge Inventory 2008). Compared to solid bridge decks, steel grid decks have seve ral advantages: they can be assembled in the factory; they are light weight; and, finally, they are easy to install. On the other hand, worn steel grid decks have high maintenance cost and provid e poor skid resistance, especially in rain. Furthermore, they provide poor riding comfort a nd high noise levels when traffic travels across the bridge. As a possible option to replace the worn out steel grid decks, the Florida Department of Transportation (FDOT) is investigating the possibility of usi ng glass fiber reinforced polymer (GFRP) decks. GFRP decks have the potential to provide a so lid riding surface, addressing the noise and stopping distance concerns of worn steel grids. GFRP decks weigh significantly less than conventional decks approximately 80% less than a deteriorated concrete bridge (Alampalli et al. 2002). The subsequent reduction in dead load increa ses the allowable live load capacity of the bridge without significant repa ir work to the existing superstr ucture and substructure, thus lengthening its service life. GFRP deck panels can be designed and manufactured to meet weight and dimensional requirement of a bridge, allowing direct replacement of steel grid decks. GFRP panels described in this thesis were premanufactured and were cut in suitable length and width. All the panels were numbered and thei r designated locations on the bridge were predefined. Panels were transported to the bri dge site and individual panels were assembled together on the bridge, following a constructi on sequence. A detailed discussion of the 14

PAGE 15

construction sequence is described la ter on in this thesis. Addi tionally, GFRP deck installation time is very short compared to that of conve ntional bridges, thereby shortening the required traffic re-routing time GFRP bridge decks are relatively new to the bridge industry. The first U.S. all-composite (FRP) vehicular public bridge was placed in service in December,1996 on No Name Creek (NNC) in Russell Kansas (MDA 2000). It was a twolane bridge, 27 ft wide, bridge has a clear span of 2l ft. and 3 in. and is constructed of three fiberglass sandwich panels measuring 23 ft.3 in. long and 9 ft. wide. The entire installation re quired one and a half days from start to finish, demonstrating the simplicity of this type of construction (Plunkett 1997). There has been continuous research on the us e of FRP bridge decks since their inception, but there are no well-adapted desi gn guidelines, nor structural an alysis procedures. A primary concern for GFRP deck systems is th eir durability and field performance. To investigate the performance of this deck type, IBRC funding was used by the FDOT to install a GFRP deck on bridge number 930338 over Hi llsboro canal in Belle Glade, Florida. The canal crossing superstructure was originally constr ucted of steel stringers with steel grid riding surface and was intended to be moveable. The objective of the IBRC study was to investigate the short and long-term field performance of the relatively new deck system. This was accomplished with the combination of long-term m onitoring and periodic lo ad tests. The load tests provided information on the behavior of an actual installation under tr uck loading. Periodic load tests combined with the monitoring data will allow estimates of truck frequency and weight carried by the FRP deck during the monitoring peri od. The selected bridge is on a main route from sugar cane fields to proces sing plants and carries signific ant unpermitted truck loads during 15

PAGE 16

16 the harvest season. This thesis presents the inst rumentation, procedures, a nd results of the first load test conducted on the bridge deck in October 2009. Objectives The objective of this research study was to calculate the distribu tion of the tire loads between the webs of the GFRP Bridge deck, set up a base line for the strain and deflection for the long term monitoring. It was also envi saged to calculate the flexural contribution (composite action) of the top plate in bending of the bridge deck. Ot her objectives of this research were to compare the field test results w ith the lab test data and calibrate the truck wheel load to strain for monitoring. This was accomplis hed by conducting a field load test on the GFRP bridge deck immediately after the installation of the deck. Scope of Work Field load test in October 2009 A field load test was conducted on the GFRP de ck immediately after the installation of the deck. Purpose of this test was to set up a base line of strain and deflection for the monitoring. Distribution of tire loads Calculation of distribution of tires loads between the webs of the GFRP deck. Load distribution was calculated utiliz ing flexural and shear strain. Calibration of wheel load to strain for load monitoring Truck wheel load was calibrated to strain, this data will be utilized for the conversion of the monitoring strain into wheel load. Composite action Flexural contribution of top GFRP plate was determined by calculating the composite action between the top and botto m section of the GFRP deck. Existing steel superstructure Performance of the existing steel girder syst ems were evaluated by measuring the strains and deflection for the exis ting steel superstructure.

PAGE 17

CHAP TER 2 EXISTING BRIDGE CONFIGURATION The bridge selected for GFRP deck replacem ent is located in Belle Glade, Florida ( Figure 2-1 A). Bridge No. 930338 is located on North Main Street and crosses ove r the Hillsboro Canal ( Figure 2-1 C) carrying five lanes of traffic. There are two northbound and two southbound lanes with a northbound left-turn lane. A m edian separates th e north and southbound lanes and sidewalks are located on each side. The bridge is situated between two intersections that have traffic signals. The north end is at the intersec tion of N. Main Street and E. Lake Road. The south end is at the intersection of N. Main Street and E. Canal Street South. The superstructure crosses the canal with thr ee short spans. Cast-i n-place concrete flat slabs make up the outside two spans of 18 ft; steel grid deck supported by structural steel framing makes up the main span ( Figure 2-2 ). Main span, which has steel grid deck, is 36 ft long. Heavy traffic occurs during the sugar ca ne harvesting season (f rom late-October through mid-April). It has been observed that the suga rcane loaded trucks trav el in the two north bound lanes noted as lane 1 and lane 2 in Figure 2-1 B. Figure 2-3 shows the local damage sustained by the steel grid deck and associated repairs using st eel p lates. GFRP deck was used to replace the steel grid deck. This grid deck was replaced by the GFRP Deck which is the focus of this study. The original steel grid deck br idge was constructed in the 1 950s as a lift bridge and was intended to allow boat traffic through the canal by using cranes to lift sections of the steel framing and grid out to allow pa ssage of the marine traffic. Figure 2-4 shows the steel superstructure fram ing plan. W24x68 steel gird ers provide the main superstructure support and are spaced at approximately 4-ft center-to-center. Girders were assembled into frames using intermediate and end diaphragms fully welded to the girders. This forms a rigid frame that can be lifted off of the substructure as a unit to allo w passage of marine traffic. To the authors 17

PAGE 18

knowledge, however this has not been done. Tra ffic and pedestrian barriers are supported by transverse m embers that are integrated with the girders under each sidewalk. 18

PAGE 19

Miami Orlando Belle Glade A B Lane 1 Lane 2 Lane 3 Median Lane 4 Lane 5 Northbound Southbound Canal Pier 39 ft Steel Grid with structural steel framing Pier 15 ft CIP concrete deck Abutment Abutment Canal 15 ft CIP concrete deck 4'-10" 4'-10" 6 27 '37 '-6" 80 '-2"Side Walk Side Walk N C Figure 2-1. Existing bridge. A) Location, B) Aerial photo (Source: www.maps.google.com Last accessed August 2009) C) Detailed site plan. 19

PAGE 20

20 Figure 2-2. Elevation view of main span (Credit: Abhay P. Singh). Figure 2-3. Damaged and repaired existing steel grid deck (Credit: Abhay P. Singh).

PAGE 21

Figure 2-4. Existing frami ng layout for lift out span. Figure 2-5. Cross sectional view of lift out span. 21

PAGE 22

CHAP TER 3 GFRP DECK DESIGN Deck Design Figure 3-1 shows the pultruded glass fiber reinfo rced polym er (GFRP) composite deck system used to replace steel grid system. E Gla ss fibers and isopolyester resin were used as the building blocks of the composite deck. Exact fi ber lay up and resin mixture was confidential and not available. Deck panels were manufactured in widths of approximately 2.5 ft and were supplied in two parts. The bottom portion of th e deck was composed of a 0.5-in. thick panel pultruded integrally with four I-shaped webs. The bottom plat e was thickened locally near each web to match the top flange of the webs. Layout of the bottom GFRP panels is presented in Figure 3-3 Span of the bottom panels is parallel to the abutment, bottom GFRP panels span between the steel gi rders as shown in Figure 3-4 A. 0.5-in. thick GFRP plate was attached to the top flanges of the bottom deck panels using 1.75 -in. long self tapping mechanical fasteners at a maximum spacing of 12 in. Top panels were also pultruded and then cut to fit the deck plan. Adjoining bottom panels were connected by fastening the protruding portion of the bottom panel to that of the adjacent panel with mechanical fasteners. Top GFRP pa nels were generally 35in. to 48-in. wide and were placed perpendicular to the direction of the bottom pane l so they spanned between the webs. Top panels laid on the bridge parallel to the curb as shown in Figure 3-4 B. Figure 3-5 shows plan layout of the top GFRP s heet. Deck Installation Deck replacement was carried out under a c onstruction contract w ith FDOT District 4, which included roadway resurfacing in addition to the deck replacement. To accommodate the deck replacement, traffic was routed around the bri dge to an adjacent bridge located a few blocks 22

PAGE 23

to the South. This section describes the constr uction sequence used to re move the existing deck and install the new GFRP deck. Existing s teel grid ( Figure 3-2 ) was removed from the superstr ucture. To ensure that the finished wearing surface of the new deck aligne d with the remainder of the bridge deck, it was necessary to provide a la yer of leveling grout between the top flange of the steel girders and the soffit of the GFRP deck. Leveling grout pads were poured using the formwork system shown in Figure 3-6 Formwork was placed such that it created a nom inal 0.5-in. gap for the grout to fill. This gap was varied as needed to accommodate the range of (0-0.5 in.) construction tolerances. Installation of the deck be gan with placement of the bo ttom panels on the leveling formwork ( Figure 3-7 ) perpendicular to the existing steel b eam s. The bottom panels had already been manufactured and cut to length and were st ored on site. Each piece was custom fit to a particular location within the bridge deck. As bottom panels were placed, they were mechanically connected using the protruding bottom deck flanges. Figure 3-8 A shows the details of the transition be tween the GFRP deck and the concrete deck on the approach sp ans. To accommodate this transition, cast-in-p lace concrete was placed over the end of the structural steel girder frames (visible in Figure 3-8 B). The edge GFRP panel was used as a stay-in-p lace form for the concrete by removing the top flanges of the three outside panel webs. Steel reinforcement was threaded through holes drille d in the webs and welded to the existing end plate on the abutment. GFRP deck was connected to the existing steel stringers with welded headed studs. Holes were drilled through the bottom FRP deck panels to accommodate the steel studs. Studs were then welded to the top flange of the existing girder through the hole in the GFRP deck ( Figure 3-8 B). Foam dams were placed adjacent to the studs to retain the grou t. Grout ( Figure 3-9 ) was 23

PAGE 24

24 then poured into the pockets fl owed through the hole and filled the space between the deck and top flange of the steel girders. When this sp ace was full additional grou t was placed to encase the welded headed stud. Longer studs were weld ed to the existing steel beams to anchor the median to the bridge deck (Figure 3-10 ). Sim ilar to the bottom panels, top GFRP sheets ( Figure 3-11 ) had already been m anufactured and cut to length and were stored on site before construction began. Top GFRP sheets were laid over the top of the bottom pane ls and attached with mechanical fastener provided by the manufacturer. Af ter installation of all top sheets, the existing median was re attached to the deck using the median anchors. After the installation of the median, a 0.5-in. thick overlay of polymer concrete ( Figure 3-12 ) was placed on the top GFRP panels to create the traffic wearing surface. Figure 3-13 shows the completed deck system open to traffic.

PAGE 25

3 sp @ 8" = 2'-0" 3.5" 3.5" 4.5" 0.5" self-tapping fasteners 4" 4" 0.5" A B Figure 3-1. GFRP deck configur ation (A) typical section (B) si ngle panel section without top plate (Credit: Abhay P. Singh). Figure 3-2. Existing steel grid deck (Credit: Abhay P. Singh). 25

PAGE 26

Figure 3-3. GFRP bottom section panel layout. 26

PAGE 27

A B Figure 3-4. Layout of GFRP panels A) bottom panels spans parallel to abutment B) t op panels span parallel to curb(Credit: Dr. Dan Richards). 27

PAGE 28

28 Figure 3-5. Top GFRP section layout.

PAGE 29

Figure 3-6. Formwork for grout pa ds (Credit: Dr. Dan Richards). A B Figure 3-7. Installation of bottom GFRP panels A) Lifting of single panel B) Connected bottom panels (Credit: Abhay P.Singh). 29

PAGE 30

A B Figure 3-8. Transition between GFRP deck and concrete deck A) Reinforcement for cast-inplace concrete B) Installation of welded headed stud (Credit: Dr. Dan Richards). Figure 3-9. Grout pockets being poured at each stud (Credit: Dr. Dan Richards). 30

PAGE 31

Figure 3-10. Median anchors (Credit: Dr. Dan Richards). Figure 3-11. Top GFRP panels (Credit: Dr. Dan Richards). Figure 3-12. Placement of polymer concrete wearing surface (Credit: Dr. Dan Richards). 31

PAGE 32

32 Figure 3-13. Bridge in service after construction (Credit: Dr. Dan Richards).

PAGE 33

CHAP TER 4 INSTRUMENTATION AND DATA ACQUISITION Instrumentation installed on the bridge was in tended to serve two purposes. One was to acquire strain and deflection data during the bridge load tests. The other was to collect strain, deflection and thermal data during the monitoring period of the bridge deck under actual traffic conditions. Instrumentation was placed solely on the superstructure (steel girder system and GFRP deck) for load tests and monitoring. S ubstructure behavior was not anticipated to significantly affect the behavior of the bridge under either load tests or actual traffic loads. Approach Two extreme northbound lanes were instrumented because these two lanes carry numerous heavily loaded sugarcane trucks during the sugarcane harvesting season. Deep pavement ruts and steel grid damage are an indication of the heavy loads transported in these lanes ( Figure 4-1 ). Bonded strain gages were applied to the GFRP d eck system in two locations. The webs were instrumented with strain rosettes (web gages) to measure shear strain. Uniaxial strain gages (soffit gages) were applied to the deck soffit to measure flexural strains para llel to the deck webs. These gages were generally placed directly unde r a web. Thermocouples were mounted on the GFRP deck in strategic locations to measur e the gradient through the deck thickness. Displacement gages were used to measure the ver tical deck deflection and the relative vertical deflection of the steel girders dur ing the bridge load test. Full -bridge strain (FBS) gages were mounted on the bottom flange of the structural steel girder. Due to the lack of access to the webs after the top plate was fastened in place, the web gages (strain rosettes) and thermocouples were installed prior to deck installation. Soffit gages, FBS gages on the steel beams were installed after deck installation and just prior to the bridge load test using a barge provided by FDOT District 4. 33

PAGE 34

Table 4-1 summarizes the instrumentation used for both the bridge load tests and monitoring. Most of the instrum e ntation except thermocouples was located at the midspan of the steel girders. Surface temperature measuring th ermocouples were installed on the GFRP deck panel B8 located closer to the DAQ. Thermo couples were also installed on the traffic box containing data acquisition system used for monito ring. Wires were run to the east side of the north abutment where the da ta acquisition was housed. Strain Gages Figure 4-3 shows the location of the four FBS gages th at were attached to the steel girders. These gages were full bridge strain transducers th at were bonded to the bottom flange of the steel beams. Girders 3, 5, 6, and 9 were each instru mented with FBS gages on the bottom flange to measure tensile strain. The FBS gages on Girder s 3 and 5 were expected to show significant strain when traffic was in lane 1. Similarly, Gi rders 6 and 9 were expected to have significant readings when traffic was in lane 2. All FBS ga ges were located at the mid-span of the girders and were used for both bridge load tests and monitoring: Figure 4-4 shows the mounting tabs and tab jig used for the installation of the FBS gage. The m ounting tabs were adhered to the bottom fla nge of the girder usi ng a two part epoxy. The surface of the steel girders were cleaned usi ng hand grinder, sand pape r and denatured alcohol and a 2-part epoxy was applied on the steel girders to attach the transducer tab on it. Mounting tabs and tab jig ( Figure 4-4 A) was used to install th e FBS gage to the existing steel gird er. For attaching FBS ga ges, mounting tabs were placed in the slot of the tab jig. Slots in the tab jig were perpendicular to the axis of the FBS gage. FBS gage was put on the tabs through and bolted. Mounting tabs and tab jig wa s used only for the purpose of aligning and attaching the FBS transducers to the steel girders. After the gage was assembled and bolted the 34

PAGE 35

whole assembly was adhered with epoxy. Figure 4-5 shows an installed FBS gage on the bottom flange of the steel girder. The GFRP deck was oriented to sp an parallel to the abutment and piers, which placed them at a skewed angle to the girders ( Figure 4-6 ). Panels were cut to le ngth before shipment to the site and were custom fabricated for a specific lo cation within the bridge deck. Panels B9 and B10 were selected to be instrumented due to their proximity to midspan and the FBS gages on the steel girders. Four 5-mm long, 350 ohm, uniaxial (UFLA-5350-11-5-LT)) bonded quarter bridge foil strain gages from Texas Measurement Inc (TMI) were mounted on the soffit of each of the two instrumented panels ( Figure 4-7 ), orientation was such that stra in para llel to the GFRP webs was measured. Four 5mm, 350 ohm strain rosettes (0-45-90) (UFRA-5-350-115LT) from TMI were installed on the webs of each instrumented panel with the zero direction gage oriented along the longitudinal axis of the web. While rosettes were used for the bridge test only, soffit gages were used for both the bridge test and monitoring. Figure 4-8 A shows an installed bonded gage on the sof fit of the GFRP deck while Figure 4-8 B shows an installed bond ed strain rosette on the web of the GFRP deck. Gages S1, S2, S5 and S6 were intended for vehi cles in lane 1 and gages S 3, S4, S7 and S8 for vehicles in lane 2. During the bridge test it was observed that gage S3 and S8 were not working properly. Thermocouples Figure 4-9 shows the location of the four genera l purpose Type K therm ocouples used to measure the temperature of the GFRP deck panels These thermocouples are capable to measure a temperature range of (-58 F to 392oF). The Bridge is located in Belle Glade, FL and it was expected that these thermocouples will be able to measure daily and seasonal variation of 35

PAGE 36

tem perature at bridge site. The thermocouples were arranged to provide continuous readings of the thermal gradient that develops during heating and cooling of the bridge deck. Due to lack of access of inside of the bridge deck, thermocouples were applied before deck installation. Panel B8 was chosen due to its pr oximity to the long-term monitoring equipment. Four thermocouples were distributed over the height of the section. Thermocouples (Omega Engineering Inc. model SA2F-K-K120-SMPW-CC) were installed at the top, mid-height, and bottom of the selected web. One additional thermocouple (Omega Engineering Inc. model SA2C-K-K120-SMPW-CC) was installed at the junction of the web and the top flange. The thermocouple used in this position is designed to be placed on a curved surface. One thermocouple was installed in the shade under the data acquisitions box to measure ambient site temperature. Omega model NB4-CAXL-14U02 was selected as a durable instrument suitable for measuring outdoor temperature. Displacement Gages Displacements of the GFRP deck and structur al steel girders were measured during the load test. In addition, GFRP deck displacements were monitored. Relative girder displacements in two locations were measured using the fixture shown in Figure 4-10 The frame is attached rigidly to one of the girders w ith the displacem ent gage plunger contacting the adjacent girder. Relative displacements were measured between girders 3-4 and 7-8 ( Figure 4-6 ). The GFRP deck disp lacement relative to the adjacent structural steel girders was also measured during the load test and during the monitoring period. The displacement gage was located in lane 2 between girders 5 and 6. Figure 4-11 shows the displacement gage and its support fram e, which was c-clamped to the bottom flanges of the adjacent steel girders. The displacement gage was mounted in the center of the frame to measure midspan deflections for the GFRP deck. The two displacement gages shown in the photo were us ed during the bridge 36

PAGE 37

load te st for the Acoustic Emission (AE) calibra tion and were removed following the bridge load test. The displacement gage used for mon itoring was model LD-620 manufactured by Omega Engineering Inc. and has a guided core with removable spring plunger with a range of + 1 in. Instrumentation Positions Truck positions were tracked with GPS during the bridge load test. To enable analysis of the instrument data with respect to the truck pos ition, it was necessary to establish the coordinate position of each gage. For the purposes of this report, the coordinate position of both the truck and instruments are recorded with resp ect to the coordinate axes shown in Figure 4-2 Inter section of the expansion joint (between GF RP deck span and in-situ concrete approach span) and curb on the SE side of the bridge is defined as the origin. The instrument positions were shown in Table 4-2 Girder 2 is 11 in. from the face of the curb. Sampling Rate Short spans and the relative flexibility of the GFRP deck was anticipa ted to cause the wheel effect on the strain and displacement gage s to be rather localiz ed. Consequently, the sampling rate for both the bridge load test and the long-term mon itoring were carefully considered. Due to the use of the GPS system during the br idge load tests, the truck was rolled across the bridge at a rate of 0.75-1 mph rather than positioned statically. The data acquisition system recorded both truck position and associated instrument readings at regular intervals. The sampling rate chosen for the bridge load test was 5 Hz. At the rolling rate used for the test, the truck wheel will take approximately 1.76 sec to tr averse a single GFRP panel. At a sampling rate of 5 Hz, approximately 9 scans were collected as the wheel traversed the panel, which was deemed sufficient to capture the deck behavior. 37

PAGE 38

Monitoring of actual vehicular traffic, however required a higher sam pling rate due to the traffic speeds. The local speed limit is 35 mph. Traveling at this rate, it takes only half of a second to traverse the GFRP deck. Moreover, the time to traverse a single 31-in. wide GFRP panel is approximately 0.05 sec. To ensure that the peak strain and deflection in the GFRP panel is captured, a higher sampling rate was required. The criterion establishe d for the bridge load test of a minimum of 9 data points for a single panel was used to establish the sampling rate for traffic monitoring. To ensure at least 10 data po ints were recorded on any single GFRP panel, a sampling rate of 200 Hz was selected. Table 4-3 shows the calculation of traverse time for the traffic traveling at the allowable speed lim it. Data Acquisition System In October 2009, the bridge load test wa s conducted. Following the load test, the monitoring data acquisition system was activated. The instrumentation for both the load test and monitoring was installed im mediately prior to the Oc tober 2009 load test. For the load test, the FDOT Structures Research Centers data acquisition system was used for collecting data from the instrumentation during the load test. This required that the instrument s be wired temporarily to the FDOTs DAQ. Following the load test, the wiring was then connected to the DAQ used for monitoring traffic. The CompactRio (cRIO 9104, 8 slot 3 M gate reconfigurab le chassis,) data acquisition system from National Instruments was used for this purpose. The system had a peak sampling rate of 1000 Hz and was fitted with a quarter bridge module (NI 9236) capable of handling 8 channels, a full bridge module (N I 9237) capable of handing 4 ch annels, an LVDT input module (NI 9215)capable of handling 4 channels, a nd two thermocouple modules (NI 9211) each capable of handling 4 channels. Figure 4-12 shows the cRIO and various input modules. 38

PAGE 39

Figure 4-13 shows the wiring diagram for the longterm monitoring system. The wires from each instrument were collected into a singl e bundle that ran inside a unistrut tray from east to west toward the sidewalk. The bundle exited the unistrut tray, entered a protective sleeve, and turned North following a steel girder to the pier The bundle entered a flex conduit at the pier, which runs around the pier and c onnects to rigid conduit that runs under the slab bridge span over to the abutment and eventually terminates at the traffic box. C-clamps were used to attach the flexible sleeveto the steel gird er using C-clamps. Rigid conduit attached to the concrete slab side span was used to carry the wiring bundle to the data acquisition system on shore. Power was provided by a solar panel ( Figure 4-13 C), which was installed next to the traffic box. Data have been continuously collected from a ll the sensors and stored in an external 16 GB USB drive acting as a remote server on site. Thes e data were transferred remotely to the FDOT Structures lab Tallahassee usi ng a cellular modem (RAVEN X from Sierra Wireless) and Verizon wireless data plan. 39

PAGE 40

Table 4-1. Summ ary of instrumentation for load tests and monitoring. Gage Location No. of gages Installed Full bridge strain gage Steel beams 4 After the construction of the deck Bonded quarter foil strain gage FRP deck panels 8 After the construction of the deck Deflection gage Steel beams and FRP deck 3 After the construction of the deck Bonded quarter foil strain rosette (0-45-90) FRP deck 8 x 3 Before the construction of the deck Surface temp. measuring Therm ocouple FRP deck 4 Before the construction of the deck Ambient temp. measuring Therm ocouple Traffic box 1 After the construction of the deck Table 4-2. Coordinate position of gages. Coordinates Gage Location x (in.) y (in.) SE corner 0 0 B1 Stringer 3 59 210 B2 Stringer 5 128.6 210 B3 Stringer 6 176.6 210 B4 Stringer 9 294.2 210 D1 B10 155 278 D2 Stringer 8 246.2 210 D3 Stringer 3 59 210 S1 B10 35 225 S2 B10 104.6 266 S3 B10 200.6 317 S4 B10 235.4 335 S5 B9 35 256 S6 B9 69.8 273 S7 B9 200.6 342 S8 B9 270.2 379 R1 B10 37 229 R2 B10 105 265 R3 B10 201 316 R4 B10 269 352 R5 B9 37 255 R6 B9 105 291 R7 B9 201 342 R8 B9 269 377 40

PAGE 41

Table 4-3. Sa mpling rate calculations. Span 35 ft Average instrumented panel width 31 in. Allowable speed on the bridge 35 mph (51 ft/sec) Time taken to cross the bridge 0.68 sec Time taken to cross the instrumented panel 0.05 sec 41

PAGE 42

Figure 4-1. Two northbound lanes showing truck traffic marks on the road surface (Credit: Abhay P.Singh). 42

PAGE 43

C L Bent 3 Bent 2 N X Y Origin for GPS and instrumentation Curb Expansion joint A B Figure 4-2. Origin for the GPS (A) location (B) Pictorial view (C redit: Abhay P.Singh). 43

PAGE 44

N 1 3 2 5 7 6 9 10 4 8 11FBS Gage C L Bent 3 Bent 2 C L Curb Lane 2Lane 1 B1 B2 B3 B4 Gage placed on top of bottom flange Midspan Figure 4-3. FBS gages m ounted on steel girders. A B Figure 4-4. FBS gage (A) Mounting tabs a nd tab jig (B) gage (Credit: Abhay P.Singh). 44

PAGE 45

Figure 4-5. Installed FBS gage on the steel girder (Credit: Abhay P.Singh). Figure 4-6. Location of instrume nted panels (B9 and B10) th ermocouples and displacement gages. 45

PAGE 46

Curb Lane 2 Lane 1 N 9 B9 B10 panel joints 2'-6" 2'-4" B9 B10 webs Web gage Web gage A Partial Plan Instrumented Deck Panels Section A 2.25" 8 7 6 5 4 3 2 R8 R7 R6 R5 R4 R3 R2Soffit gage S5 S6 S7 S8 S1 S2 S3S4 Soffit gage R1 Figure 4-7. Position of bonded st rain gages and rosettes on GF RP deck panels B9 and B10. A B Figure 4-8. Instrumentation on the GFRP Deck (A) Uni-axial bonded gage (B) 0-45-90 strain rosette (Credit: Abhay P.Singh). 46

PAGE 47

A B Figure 4-9. Thermocouples on the surface of the GFRP deck (A) locati on (B) Pictorial view (Credit: Abhay P.Singh). 47

PAGE 48

A B Figure 4-10. Displacement gages on the steel gi rders (A) Schematic (B) Pictorial view (Credit: Abhay P.Singh). A B Figure 4-11. LVDT on the GFRP deck (A) Sc hematic (B) Pictorial view (Credit: Abhay P.Singh). 48

PAGE 49

Figure 4-12. Monitoring DAQ (cRIO) and va rious input modules (Credit: Abhay P.Singh). 49

PAGE 50

50 A B C Figure 4-13. Location of the cRIO and instrumentation wiring (A) conduit layout (B) cRIO inside the traffic box mounted on the sign pos t. C) Solar panel next to traffic box (Credit: Abhay P.Singh).

PAGE 51

CHAP TER 5 BRIDGE LOAD TEST OCTOBER, 2009 Objectives The objectives of the load test were to m easure the GFRP deck reference strains and deflections to which future monitoring and load test results can be compared and measure the lateral load distribution in th e GFRP deck panels. It was also intended to calculate the contribution (composite action) of the top plate in the flexural stiffness of the two part GFRP deck system. Other objectives of the load test we re to compare the field test results with the lab test data, calibrate the truck wheel load to strain so that monitoring strains can be converted into the wheel loads. Lastly, it was also envisaged to evaluate the performance of the existing steel superstructure. Truck Positions and Load Levels FDOT Structures Research Center load trucks were used in the load test ( Figure 5-1 ). The trailer was designed to impose known wheel and axle loads to the bridge as a function of the num ber of blocks stacked on the trailer. Table 5-1 shows the axle loads associated with the num ber of blocks stacked on the trailer. Five truck positions (TP1 through TP5) were used to maximize the bending and shear effects in the strain gages mounted on the GFRP panels and to cover most combinations of traffic movement transversely in the two instrument ed northbound lanes. All truck positions were marked parallel to the curb by measuring the dist ance from the face of the curb to the intended position of the tires on the west side of the truck. TP1 through TP5 are shown graphically in Figure 5-2 through Figure 5-6 Figure 5-7 shows the distance from the face of the curb to th e outside of the wheel on the drivers side. 51

PAGE 52

Lines m arked from 1 through 5 on the deck are th e five truck positions. Face of the curb is located 1.5 ft away from inside of the first lane mark. TP1 and TP4 corresponded approximately to the tire marks in each lane and are considered to be the path that trucks will typi cally take when traversing the bridge. Four load steps were used for the bridge test ( Table 5-1 ). Maximum axle load applied at load step four was 35.5 kip, which resulted in a wheel load of approxim ately 18 kip. Note that the wheel load is spread over the two tandem tires on the rear trailer axle. This load was chosen to ensure that the AASHTO design service wheel load of 16 kip was reached. AASHTO LRFD (2007) section 3.6.1.2 specifies a design wheel load of 16 kip (72 kN) x 1.33 dynamic load allowance = 21.3 kip (94.7 kN). Testing was in itiated with 12 blocks and was increased in increments of 6 blocks to a maximum of 30 blocks. Test Setup The instrumentation needed for the bridge te st and monitoring were installed during the two days prior to the load test. The actual brid ge test was performed at night from 9pm to 5am to avoid causing traffic delays. The FDOT lo ad test DAQ and AE system were placed on the east sidewalk of the bridge during the test. Th e labeled instrumentation wires were hooked up to the FDOT load test DAQ. Truck position reference marks were painted on the bridge deck ( Figure 5-8 ). GPS was us ed during the load test to record the position of th e truck along with each data scan of the instruments. Figure 5-9 shows the position of the GPS do m e with respect to the truck axles. Prior to the load test GPS dome was used to take position readin gs of several reference points such as origin for GPS and instrumentatio n, ends of the truck positions and end of the marked instrumented panels on the bri dge. Accuracy of the GPS was 2 in. 52

PAGE 53

Test Proced ure Bridge test was performed in two steps, star ting with the static te st by driving the truck over the bridge and parking it at a location where it maximize the strains and deflection. Followed by static test, a slow roll test was conducted by slowly (0.75-1 mph) rolling the truck on the bridge in all the truck positio ns and for all four load levels. Step 1 Static Load Test This step was performed by stat ically positioning (non-rolling) the two trucks to determine if it was necessary to use both trucks for the entire load test or if one truck would be sufficient to capture the behavior from the bridge deck. The test was started by positi oning the truck in lane 1 at TP2 ( Figure 5-3 ) with the rear axle over the instrumented panel B9. It was necessary to adjust the truck position slightly to m aximize the strain in soffit gage S5. Strain and deflection were recorded when truck was in this position. Leavin g the first truck in lane 1, the second truck was positioned in lane 2 at TP5 ( Figure 5-6 ) with the rear axle over the instrum ented panel B9. Strain and deflection were reco rded. Leaving the second truck in lane 2, the first truck was removed from lane 1. Strain and deflection were recorded and the static test was terminated. Step 2 Rolling Load Test A single truck was rolled slowly (0.75 1 mph) across the bridge in all five truck positions as shown in Figure 5-7 For each load step, the truck was rolled through all five truck positions before m oving on to the next load step. To a llow correction for residual strain, zero readings were recorded prior to every load step. Figure 5-10 shows a flow chart of the load tes t procedure followed during the field load test. During the entire test strain and deflection from selected gages were plotted and monitored for linearity. This helped to avoid damaging the bridge during the test. 53

PAGE 54

Table 5-1. Test truck axle loads (kip). Front Tandem Rear Tandem No. of Blocks Front Axle P1 (kip) P2 (kip) P3 (kip) P4 (kip) P5 (kip) 12 11.26 10.80 18.99 18.99 18.99 18 11.58 10.00 25.53 25.53 25.53 24 11.22 11.01 30.60 30.60 30.60 30 11.36 11.95 35.49 35.49 35.49 54

PAGE 55

Figure 5-1. FDOT utility truck used for Field Load Test C L B e n t 3Bent 2 C L Curb L a n e 2 L a n e 1 FDOT Load Test Truck N Figure 5-2. Truck in position TP1. C L B e n t 3Bent 2 C L Curb L a n e 2 L a n e 1 FDOT Load Test Truck N Figure 5-3. Truck in position TP2. 55

PAGE 56

C L B e n t 3Bent 2 C L Curb L a n e 2 L a n e 1 FDOT Load Test Truck N Figure 5-4. Truck in position TP3. C L B e n t 3Bent 2 C L Curb L a n e 2 L a n e 1 FDOT Load Test Truck N Figure 5-5. Truck in position TP4. C L B e n t 3Bent 2 C L Curb L a n e 2 L a n e 1 FDOT Load Test Truck N Figure 5-6. Truck in position TP5. 56

PAGE 57

Edge of curb End of GFRP DeckB9 Instrumented PanelsB10 L a n e 2 L a n e 1 Center of median 15'-11"N 1 2 3 45 1 0 1 1 2 2 1 3 1 0 2 1 9 "2 4 6 2'-10" 2'-8" 1 6 First lane mark Figure 5-7. Truck positions for brid ge test. (Lines indicate outside edge of tires on west side of truck). Figure 5-8. Truck position re ference marks on the bridge deck (Credit: Abhay P.Singh). 57

PAGE 58

58 Figure 5-9. Location of the GPS dome and relative location of the truck axles. Rolled truck in TP1 through TP5 Recorded strain and deflection by load test DAQ at all channels Drive truck off the Bridge Record zero reading Add 6 extra Blocks on the trailer Record zero reading Monitor strain and deflection for linearity P e r f o r m a l l th e st e ps upto 3 0 B l ocks Figure 5-10. Rolling Test Flow chart.

PAGE 59

CHAP TER 6 BRIDGE TEST RESULTS OCTOBER, 2009 Overview This chapter presents the results of the first bridge test conducted af ter the installation of the GFRP deck on October 14, 2009. Initially a static load test was conducted using two load trucks positioned individually, a nd then in tandem, over the inst rumented GFRP panels. Based on the strain results from the static test, it was found that a single truck could be used for the remainder of the load test. The remainder of the testing was conducted with the single truck by rolling slowly over the bridge while recording instruments and truck position data. Utilizing strain and truck position (GPS data), strain infl uence lines were created for the soffit gages and full bridge gages for their respective locations on the bridge. Load distribution between the webs of GFRP deck was calculated using the influence lines. Deflection gage data are presented in load-displacement plots for GFRP deck. Response of existing steel girders is presented in terms of strain and deflection. Co mposite behavior between top and bottom GFRP section were calculated analytically and experi mentally. Load-strain calibration plots were created for all of the soffit gages. Finally, laborat ory test results are compared with the results of the bridge test. Static Load Behavior It was anticipated that the response of the d eck system would be su fficiently localized so that a single truck, rather than tw o, would be sufficient to capture th e behavior of the deck. This behavior was confirmed at the beginning of th e load test by using the following sequence of static truck positions using 12 blocks in each truc k. Initially a single truck was positioned in lane one at TP2 ( Figure 5-3 ) and a second truck in lane two at TP5 (Figure 5-6 ). A second set of readings was taken with a truck in T P2 but no truc k in lane two. Finally, a third set of readings was taken with a truck in TP5 but no truck in lane one. For each st atic load case, the trucks were 59

PAGE 60

m aneuvered into position with the rear trailer axle over the instrumented panels until the data acquisition indicated a maximum strain was reached at strain gage S5 for truck in lane 1 and strain gage S7 for truck in lane 2. Load test DAQ continued recording st rains and deflections at these truck positions for almost 25 30 sec. Load test DAQ had a sampling rate of 5 Hz, recording nearly 125-150 data poi nts for each static load test. Soffit gage strains were corrected for the a ppropriate zero load reference readings. Averages of the maximum corrected strains were calculated for the three static load cases ( Table 6-1 ). From Table 6-1 it can be observed that the maximum strain was recorded by soffit gage S7. W hen loaded with one truck in la ne 2(TP5), the recorded strain was 418 and when loaded with two trucks single truck in each lane (TP2+TP5), it was 442 The strain data reveals that the effect of the wheel lo ads was localized and th at the maximum difference in strain caused by an adjacent truck was 5.5 percent.. Rolling Load Behavior In general, bridge load tests were conducted by positioning a truck with a known load at some point on the bridge and taking a reading from the instrumentation. The FDOT trucks, however, had recently been equipped with global positioning systems (GPS). This allowed the truck position to be recorded by the data acquisi tion system each time a scan of the instruments was taken. Consequently, the tr uck could be rolled over the bridge whil e acquiring data and truck position. In additio n to reducing the test time, this instrumentation allowed the influence lines for strain and deflection to be gathered in an automated manner. Slow roll tests were conducted using a single truck with incr easing loads after the static load test. A single test truck was rolled slowly at the speed of 0.75 1 mph over the bridge through TP1 to TP5 for four load levels. A llowable speed limit on the bridge was 35 mph therefore regular moving traffic had some dynamic wheel load effect on the bridge. But due to 60

PAGE 61

slow rolling of the test truck, dynam ic wheel load effect of th e bridge was not captured during the first bridge test and the strains and deflections presented in this sect ion were nearly static values. Strain vs. truck positions were presented in terms of influence lines for GFRP deck and for existing steel girders. Accuracy of GPS was establishe d by carrying out a comparative study between actual truck axle distance and truck axle distance calculated from influence line plots. The effect of continuity of the bridge was pr esented in terms of positive and negative flexural strains calculated from the rolling load test data. Rolling load test data was analyzed for twenty load cases. Rolling load test data were corrected for appropriate zero load reference strain. Strain time history was plotted for six bonded gages and four FBS gages. The strain-time history plots ( Figure 6-1 ) show five peaks, which correspond to the five axles of the truck. The calculated strain was nearly zero at tim e zero and comes back to nearly zero once the last ax le moves off of the bridge, demonstrating that no residual strain was present a nd that the deck remained elastic during this load step. To track the truck movement by GPS, Easti ng (X) and Northing (Y) of the truck positions were recorded at every 0.2 sec (FDOT Load Test DAQ sampling rate was 5 Hz), corresponding strains and deflections were also recorded at every 0.2 sec. The combined GPS and instrument data were used to create strain influence lines. The GPS coordinate data were transformed so that each reading reflected the nor th-south distance from the strain gage of interest to axle P4 on the test truck. Axle P4 was selected as the refe rence axle because the strain was generally at a maximum at this location or at axle P5. From Figure 6-2 it can be observed that for gage S1 and S5, m aximum strain occurred at axle P5 while for S2 and S7, it was at axle P4. Negative x values indicate that P4 is south of the gage and positive values indicate that P4 is north of the gage. Note that the truck was traveling from south to north during data acquisition. This 61

PAGE 62

resulting inf luence lines for strain is shown in Figure 6-2 through Figure 6-3 The four plots in each graph correspond to one of the four load le vels 9.5 kip (43 kN), 13 kip (58 kN), 15.5 kip (69 kN) and 18 kip (80 kN) used in the bridge test. The x-axis in each graph reflects the position of P4 relative to the strain in the gage. When x is zero, then P4 is directly ove r the strain gage and is causing a maximum strain. When P4 is approximately 45 ft (13.7 m) south of the gage then the front axle (P1) is directly over the gage. Consequently, the mirrored shape of the truck ta kes form in the plots with the five peaks representing the five truck axles. The GPS locations of the peak strain values associated with each of the axles ( Figure 6-4 A) compare very well with the di m ensions of the truck shown in Figure 6-4 B. Maximum error in the m easure ment of truck position is 2%, this is also an indicator of the accuracy of GPS. Gages S1, S2, S5 and S7 show pos itive strains indicating tension ( Figure 6-2 ) in the bottom or positive bending moment. Figure 5-2 and Figure 5-6 show the orientation and wheel path for TP1 and TP5 respectively in lane one an d lane two. F or TP1, the forward motion of the truck carries the left wheel line over gage S2, whic h is located at the midspan of the GFRP panel. At this location, the girders are spaced at 4 ft (1.22 m center to center). As the wheels move over the gage, the resulting moment is positive. As the wheel continues its northerly motion to the next GFRP panel, a peak negative (compressive) strain is noted in gage S6 ( Figure 6-3 B), which is positioned in the short span (1.8 ft [0.55 m]) adjacen t to the pane l over which the wheel passes. This reflects the negative moment generated by the continuity of the panel over the steel girder. Figure 6-2 D and Figure 6-3 A indicate that the deck panels are exhibiting sim ilar behavior in lane 2 with the truck in position TP5 ( Figure 5-6 ). 62

PAGE 63

Figure 6-7 shows the influence lines for gage S5 (positive bending) and gage S6 (negative bending) for TP1 ( Figure 5-2 ). Figure 6-7 shows the relative location of positive and negative gages. From these influence lines, it can be observed that depending on the wheel path, strain changes sig n. Change of strain sign indicates that the deck transitions from positive to negative bending. This is also an indicator of the continuity of the brid ge deck. Maximum positive strain (S5) is 751 while maximum negative strain is 61 for the same load and same truck path (TP1). This shows that negative bending is not that significant (about 10% of the positive bending) but it goes through the cycle of positive to negative bending. Similar behavior can be observed for TP5 ( Figure 5-6 ) for gages S4 and S7 ( Figure 6-8 through Figure 6-9 ) Maxim um strain (751 ) is recorded by gage S5 during the entire bridge test corresponding to the maximum wheel load (18 kip) used during the test. From influence line plots, it can be observed that the strain gage data for each load le vel is very consistent except at the wheel load location, where a di fference in strain is observed as the load level changes. Figure 6-10 A shows that as the load level increases, peak strain increases in gage S7, while Figure 6-10 B shows that peak strain reco rded by gage S7 decreases as th e load level increases. This was caused due to the fact that the front of the trailer is lifting as more blocks are being added at the back of the trailer. The influence lines also indicate that the effect of the wheel load on panel st rain is quite localized. Figure 6-10 shows the influence line within the immediate vicinity of strain gage S7. Figure 6-10 A shows axle P4 (one of the rear axles) as the reference position on the truck while Figure 6-10 B shows axle P1 (front axle) as the refe rence position on the truck. The front axle of the truck has two wheels per axle, while the rear axles have four wheels per axle. Consequently, the wheel lo ad for the front axle is spread over a single tire patch while the wheel load for the 63

PAGE 64

rear axle is s pread over a dual tire patch. In either case, the stra in decreases rapidly as the tire moves away from the gage. For example, strain d ecreased to nearly half of the peak strain when P4 had moved to the adjacent web, which was 8 in (203 mm) away. As the wheel moved to the next web (at 16 in [406 mm]) the strain dropped to 27% of the peak strain Strain dropped to 10% of its original value as the wheel moved to 32 in (813 mm) where the wheel had moved to the adjacent GFRP deck panel. Similar be havior was observed in remaining bonded gages recording positive bending. Load Displacement Load displacement plot was created for th e GFRP deck. Maximu m measured deflection of GFRP deck was compared with the se rvice limit deflection recommended by AASHTO LRFD. Deflection of steel girder s is also presented in this section. Deflection of GFRP deck was measured with respect to the steel girders and relative deflection of the deck was presented in terms of the load displacement plot. Figure 6-11 shows the load displacement for th e deflection gage D1, which was located in lane 2 and m easured the displ acement of the deck relative to the steel stringers. The plot shows that significant displacement was produced only when the truck was in position TP5. Such behavior confirms the appa rent localized wheel load effect observed in the strain data. Maximum relative deflection produced during the load test was 0.09 in. (2.3 mm) for the maximum wheel load of 18 kip (80 kN) used during the test. The load displacement plot is linear indicating that th e GFRP deck material never goes in to non-linear range of its material behavior. AASHTO LRFD (2007) Section 2.5.2.6.2 specifi es a service limit for deflection as span/800 for steel, aluminum, and or concrete. The span of the GFRP deck is 4 ft. Maximum 64

PAGE 65

deflection m easured during the bridge test was 0.09 in 50% more than the service limit deflection recommended by AASHTO (0.06 in.). Deflection recommendation provi ded by AASHTO was developed for steel and reinforced concrete bridges. However AASHTO made this criterion to be mandatory for the orthotropic bridge decks. For FRP bridge decks, the defl ection, not yet addressed in current codes, is tentatively recommended to be within the limit of Span/800 by FHWA (2002). Deflection criteria are expected to be re quired to FRP bridge decks, because deflection can be significant due to their low stiffness. It was also reco mmended that deflection limit on FRP bridge deck should be in the range of span / 600 and span / 800 (Park et al. 2004). Span / 600 for the GFRP deck used in this thesis was 0.08 in.., which is close to the measured value in the field. Figure 6-12 shows the relative midspan displaceme nt of two adjacen t steel girders. Deflection of the steel girders was measured to ca lculate the relative deflection of the frame (four steel girders joined together with end diaphrag m forming a frame). Deflection gages D2 and D3 were located at the midspan of the bridge, installe d on the steel girders. D2 is located in lane 2 and D3 is located in lane 1. From Figure 6-12 A, it can be observed that when truck is in lane 1, m aximum downward deflection of 0.03 in was reco rded at D3, a very small upward deflection of 0.007 in. was recorded at D2. Upward deflection at D2 indicates lifting up of the deck in lane 2 while truck in lane 1. Similar behavior was observed by D3. Figure 6-12 show displacement tim e history for deflection gages D2 and D3. Th ese plots are for the maximum wheel load (18 kip) used during the bridge test. Maximum rela tive deflection of the st eel girders is 0.03 in. (0.76 mm) for TP1 and 0.05 in. (1.27 mm) for TP5. Maximum relative deflec tion of steel girders is almost half (55%) of the maximum GFRP deflection recorded for 18 kip of wheel load. 65

PAGE 66

Composite Behavior Belle Glade bridge superstructure was a three part system It was made up of steel girders, bottom and top GFRP panels. Bridge was c onstructed by installing pre-manufactured GFRP deck panels on the existing steel girders. C onstruction was started by installing bottom GFRP deck panels on the steel girders. Bottom panels were attached to the steel girders by steel studs. Steel studs were welded to the steel girders through the holes made into the bottom plate of the bottom GFRP deck panels. To provide better fixity in long term, grout was poured around the studs. Top GFRP sheets were attached to the bottom GFRP panels by mechanical fasteners. Construction procedure describe d indicates that some degree of composite action is possible between the steel girders and bottom GFRP deck and between top GFRP sheets and bottom GFRP deck panels. The degree of composite action between top and bottom GFRP panels was determined by comparing measured and theoretical elastic neutral axes. Theoretical neutral axis was determined by calculating the geometric center of a transformed section fo r a typical GFRP deck panel. Typical GFRP deck panel includes bo ttom plate, four I-beams and top plate. Transformed section properties were calculated by transforming the different elastic modulus into one reference elastic modulus. Bottom GF RP plate was considered as a reference for calculation of the transfor med section properties. Figure 6-13 shows the elastic modulus map used for the calculation of the transf ormed sect ion properties. Modulus map was provided by the deck manufacturer. The theoretical neutral axis location of the GFRP deck was determined both considering and ignoring the top plate. In addition, the neut ral axis location was also determined assuming a constant modulus of elasticity over the entire cross section. Thes e locations were expected to bound the measured location and to provide a relative indication of the contribution of the top 66

PAGE 67

plate to the load capacity of th e deck. Measured neutral axis was determined using the soffit gage-located at the extreme botto m fiber-and the strain from the longitudinal leg of the rosette gage-located at the mid height of the web. Ma ximum strain values obtained from soffit and rosette gages were plotted against the depth of th e section to determine the measured neutral axis location. Figure 6-14 and Figure 6-15 show location of different types of neutral axes for selected soffit gages and rosettes. .Soffit gage S1, S2, S5, S7 and respective rose ttes R1, R2, R5 and R7 were used to provide a neutral axis location. Soffit gages S1, S2, S5 were located in lane 1 and were m aximized by TP1 ( Figure 5-2 ). TP5 ( Figure 5-6 ) maximized the strain in soffit ga ge S7 in lane 2. The strain values show n in the figures are the maximum strains measured in the soffit gages along with the corresponding strain measured in the rosette gage s. When considering ga ge S1 and rosette R1, for example, maximum strain was found from strain time history plot for S1 for truck position TP1. R1 strain corresponding to the same time and truck position at which the soffit gage strain was maximum, selected from the strain time history. A similar procedure was used for determining the measured neutral axis for maxi mum rosette strain and corresponding strain in soffit gages. Maximum soffit gage strain and corresponding rosette strain were plotted against the depth of the section. Inters ection of this line a nd the reference line fo r zero strain is the measured neutral axis location. Figure 6-14 and Figure 6-15 present the measured neutral axis plotted for four selected soffit gages and rosettes. From these figures, it can be observed that data is very consistent and the location of the measured neutral axis remains nearly at the same location for all four load levels. Consistent nature of these plots indicate that for all the load level GFRP deck material always remains in its linear-elastic range of the material behavior. From Figure 6-16 it can be observed that measured 67

PAGE 68

neutral axis was consistently lo cated alm ost 0.7 in below the calculated neutral axis for entire section with top GFRP sheet but located very close (0.2 in.) to the calculated neutral axis ignoring top sheet. Measured neutra l axis aligns with the calculat ed neutral axis for the section without the top plate and considering uniform el astic modulus throughout the section. From this observation it can be stated that the contribution of top GFRP sheet in flexure of the deck is insignificant and self tapping screws do not en force strain compatibility and only partial composite action is possible. Self tapping screws were used for attaching the top plate to the bottom GFRP deck panels. From measured neutra l axis it can also be observed that significant portion of the web is under tension. Figure 6-16 presents a summary of the measured and calculated neutral axes plotted for selected gages. Average of the neutral axes wa s calcu lated for all the ga ges and plotted in this figure. It can be observed from this figure that measured neutral axis coincide with the calculated neutral axis for section ignoring top plate and constant elas tic modulus throughout the section. In conclusion the top GFRP sheet does not act in a co mposite manner with the bottom portion of the deck. Performance of the Existing Steel Superstructure W 24x68 steel girders form the main span of the structure; span parall el with the direction of traffic; and are supported by the piers. The steel girders are at 28.73 degree skew angle from the piers. Deck span direction is parallel to th e piers resulting in the same skew with the steel girders. Steel girders run at skew angle of 28.73 degree from the piers ( Figure 2-4 ) and support the GFRP deck. The girders were assem bled into frames using intermediate and end diaphragms fully welded to the girders. This forms a rigid fr ame that can be lifted off of the substructure as a unit to allow passage of marine traffic. 68

PAGE 69

To evaluate the perform ance of the existing steel superstructure, girders were instrumented using full bridge strain gages (FBS). This in strumentation was used for both load test and monitoring. Strain data from these gages combined with GPS data were used to create strain vs. truck positions for steel girders. Strain vs. tr uck position plots were refe rred as the influence lines and presented in this section. All the FBS gages were located at the mid span of the bridge. FBS ga ges were installed at the top portion of the bottom flange of the steel girders. Using strain compatibility and assuming no composite action between steel girders and GFRP deck, strain data at FBS gage location were converted to the strain at the extreme bottom fiber of the steel girder as shown in Figure 6-17 Calculated s trains at the extreme bottom fiber were plotted against the GPS data to create strain influence lines for the steel girders. Th e GPS coordinate data we re transformed so that each reading reflected the north-south distance from th e full bridge gage of interest to axle P5 on the test truck. Axle P5 was selected as the refe rence axle because the strain is generally at a maximum at this location. Negative x values indi cate that P5 is south of the gage and positive values indicate that P5 is north of the gage. Note that the truck was traveling from south to north during data acquisition. This resulting influence lines for strain is shown in Figure 6-18 The four plots in each graph correspond to one of th e four load levels 19 kip (86 kN), 26 kip (116 kN), 31 kip (138 kN) and 36 kip (160 kN) used in the bridge test. The x-axis in each graph reflects the position of P5 relative to the strain at the gage locati on. When x is zero, then P5 is directly over the strain gage a nd is causing a maximum strain. Using the load test data, load stress plots we re created for the steel girders. Stresses were calculated at the extreme bottom fiber of the st eel girder using strains at this location and modulus of elasticity for steel (29000 ksi). The peak values of stress from each load case were 69

PAGE 70

plotted against the corresponding axle load to create the load stress plots ( Figure 6-30 ). X axis reflects calcu lated stresses and Y axis is the combined weight of the rear axle P4 and P5. It has been observed that these two rear axles together produce maximum strain in the steel girder and other axles were off the bridge when rear axle is at the mid span location. Each graph has five plots indicating fi ve truck positions. Plots for each truck position have four points that corresponds to four load levels used during the bridge test. Calcul ated maximum stress was 8.2 ksi for the FBS gage B1. B1 is located in la ne 1 and activated by TP2. Yield stress of the existing steel superstructure was 36 ksi. Theref ore a safety factor of 4.4 was found from this research study. Field Test versus Laboratory Test This section presents a comparison between field and laboratory test, conducted on the same deck. Field test was conducted with the GFRP deck in place and rolling the test truck across the bridge for different load levels. Lab test was conducted on a piece of the same deck in the lab environment. Comparison was carried out in terms of the applic ation of loading, support conditions and the test results. Vy as et al. (2006) performed a lab test on the similar GFRP deck at FDOT Structures Lab Tallahass ee. A loading pad as shown in Figure 6-19 A was used for the lab test. Loading pad dim ensions used during the Lab test were. 0.25 m x 0.51 m, which conforms to AASHTO Section 3.6.1.2.5. Steel girders spaced at 4 ft center to center were used as the support for the lab test. This simulates the actual girder spacing in bridge configuration. For the field test, a FDOT test truck (Figure 6-19 B) was used from FDOT Structures Lab Tallahassee. Test truck used for the fiel d test closely resem b les AASHTO design truck. The maximum strain observed was 751 for gage S5 with the tr uck at TP1 (lane 1) and 660 for gage S7 with the truck at TP5 (lane 2) during the field test. These strains were recorded with the trucks loaded to their highest level of 30 blocks, which translates to a rear 70

PAGE 71

wheel load of 18 kip (80 kN). Vyas et al. (2009) m easured service level strains of 700 under a load of 20 kip (90 kN) during the laboratory test. AASHT O LRFD (2007) Section 3.6.1.2 specifies a design wheel load of 16 kip (72 kN) x 1.33 dynamic load allowance = 21.3 kip (94.7 kN). Failure strain measured during th e lab test was between 4000-6000 Load Strain and Load Displacement ( Figure 6-21 ) remains linear elastic up to the peak load of 83 kip (370.5 kN) at which point a loud noise was heard, ac companied by a load drop of approximately 25% (Vyas et al. 2009). The strain readings from directly under the loading pad (S10 in Figure 6-21 A) decreas ed slightly after the load peak, indi cating a delamination in the panel. Specimen eventually failed from web buckling at a load of 90 kip (396.8 kN). Load Distribution GFRP deck was constructed as a series of bottom GFRP panels and top GFRP sheets. Individual bottom and top panels were connected together by mechan ical fasteners. Top plate is 0.5 in. thick and bottom panel has a depth of 4.5 in. A single bottom panel typically has four webs. To determine the distribution of the wh eel load between the webs of the GFRP deck, distribution factors were calculate d. Distribution factors were cal culated by utilizing the soffit gage and rosette data. Soffit gages were located at the extreme bottom fiber of GFRP deck, rosettes were located at the mid height of th e web. Flexural strains were calculated from the soffit gage data and shear strains were calculate d from rosette data. Di stribution factors were based on the flexural strains calculated from so ffit gage data, distributi on factors were based on the shear strains calculated from rosette gages. Procedure for th e calculation of the distribution factors and comparison of the distribution factors for soffit gages a nd rosettes were discussed in this section. 71

PAGE 72

Distribution Factors Based on Flexural Strain Distribu tion factors were obtained for the soffit gages recording positive bending. Distribution factors were calculated for the maximum load level (wheel load of 18 kip) used during the test. Truck positions were chosen such that it maximizes the strain in the corresponding soffit gage. Figure 6-22 shows typical influence line and instrumented panels used for the calcu lation of distribution factors. Distribution factors were calculated for the selected soffit gages recording positive bending. Influence lines were utilized for the calculation of distri bution factors. Influence lines were selected from Figure 6-2 for respective gages. In this figure there are four plots corresponding to the four load levels used dur ing the load test. For distribution factor calculation, only plot corresponding to m aximum load level was us ed. Influence lines were corrected for the shift (distance of the peak fr om zero) between the peak and location of zero (gage location). This ensures that peak aligns with zero. For the calcul ation of the distribution factors only a part of these shifte d influence lines in the vicinity of axle P4 or P5 (depending on the axle where strain maximizes) were utilized. These shifted and partial influence lines were referenced as the modified in fluence lines and presented in Figure 6-23 These modified influence lines were utilized for the calculation of distribution factors for respective gages. For exam ple, for the calculation of distribution factor for gage S1, influence line from Figure 6-2 A was considered from zero to 135 in. (3.43 m). Th is includes 16 webs (including the web at gage location) in the calculations of the distribution factor. The webs of the GFRP deck were 8 in. (203.2 mm) apart, due to the sk ew of the bridge, 9 in. dist ance between the webs for the calculations of the distribution factor was considered. A mirror image of the influence line was created on the negative side of the x-axis, assu ming same influence line continues on both side of the gage. This was an assumption because from Figure 6-2 A, it can be observed that there is a 72

PAGE 73

significant influence of the adjacent axle (P 4) on rear axle P5. Sim ilar methodology was followed for the other three gages. Distribution factors were obtai ned for the instrumented web by calculating the value of strain at this web and dividing it by the sum of strains calculated at a web on the either side of the instrumented web and the stra in at the instrumented web. Sixteen webs were considered on either side of the instrumented web for this purpos e. Strain almost drops to 5% of its peak value once wheel load is 16 webs away from the instrumented web. Table 6-2 presents the distribution factors for four selected soffit gages. Average distribution factor was found 0.24. Distribution Factors Based on Shear Strain To com pare the distribution factors obtained from soffit gage data, distribution factors were also calculated from strain rosettes. Rosettes used for the bridge test were 0-45-90 degree rosette. Four rosettes R1, R2, R5 and R7 were chosen corresponding to soffit gages S1, S2, S5 and S7 for the purpose of comparison. Rosettes R1, R2 and R5 were located in lane 1 while R7 was located in lane 2. Influen ce lines were created using GPS da ta and calculated shear strains as shown in Figure 6-5 Shear strains were calculated fo r the selected rosettes using Mohrs circle form ula as given in Equation 1. )(*2cab xy Equation 1 The combined GPS and shear strain data were used to create shear strain influence lines. The GPS coordinate data were transformed so that each reading refl ected the north-south distance from the strain gage of interest to axle P5 on the test truck, this is shown on the xaxis. The y-axis represents calculated shear strain. In general R1, R2 and R5 measured the minimum shear strain at axle P5. Rosette R7, on the othe r hand, measured a minimum shear strain at axle P4. Unlike flexural strain, shear strain transitions from negative to positive values depending on 73

PAGE 74

wheel location Figure 6-24 As wheel load moves past the rosette as shown in Figure 6-25 shear force chang es sign. To compare the distribution of load, influence lines were plotted together for the soffit and web gages as shown in Figure 6-26 It can be observed from this figure that soffit gage data show a wider distribution of strain com pare to web gage (rosette) data. For the calculation of dist ribution factors, influence lines were selected from Figure 6-5 In this figure, the four plots correspond to the fo u r load levels used during the load test. For distribution factor calculation, only the plot corresponding to the maximum load level was used. Similar to soffit gage, influence lines were corr ected for the shift between the peak and location of zero (gage location). This ensu res that peak aligns with zero. Shear strain influence lines alter sign from negative to positive near at any axle. For the purpose of the calculation of distribution factors, influence li nes on the negative side of y-axis are considered. For the calculation of the distribution f actors similar approach as soffit gages were adopted. Only difference is that only three webs were considered on either side of the instrumented web for this purpose, while sixteen webs were considered fo r the soffit gage because soffit gage influence lines have wider distribution of strains compare to the rosettes. Shear strain almost drops to 2% of its peak value once the wheel load is thr ee webs away from the instrumented web. Table 6-2 presents the distribution factors for four selected rosettes. Th e averag e distribution factor was 0.38. Comparison of Distribution Factors An average distribution factor of 0.24 was calculated from the soffit gages compared to 0.38 for the web (rosette) gages. Soffit gages ha ve lower distribution factors compared to web gages due to load distribution though the bottom GFRP plate (0.69 in.) to the extreme bottom 74

PAGE 75

fiber of the GFRP deck. Table 6-2 presents a comparison of the distribution factors for soffit gages and strain rosettes. Load Strain Calibration Curves Strains generated in GFRP deck were plotted against the load level used during the bridge test. These plots were the indicator of the loca l wheel load response of the GFRP deck. These plots also indicate that both mate rial (GFRP and Steel) remained in the linear elastic range of the material behavior. Load stra in calibration curve will be used for converting monitoring strains into wheel load. Due to extremely local wheel lo ad response of the bridge deck, wheel load were used for producing the load stra in calibration for the GFRP deck. The peak values of strains from each load case were plotted ag ainst the corresponding wheel load to create th e load strain plots ( Figure 6-28 and Figure 6-29 ) for six bonded gages (f our gages recording positive bending and two r ecording negative bending). Due to the local nature of the deck behavior, wheel loads were used instead of axle loads. Each graph has five plots indicating five truck positions. Plots for each truck position have four points that corresponds to four load levels used during the bridge test. The local nature of the behavior of the deck under the wheel load is revealed in load strain plots when considering the transverse truc k position within the traffic lane. For example, consider the variation in strain at S2 as the truck position moves from TP1 to TP5. The strain recorded for TP1 is nearly six times greater than that of any other truck positions. Similarly, S7 records significantly higher strains for TP5 than for any of the other truck positions. To a lesser extent, a similar effect is not ed in the compressive strain recorded by gages S4 and S6 ( Figure 6-29). 75

PAGE 76

The load s train calibration curves remain almo st linear for all load levels, indicating that during the test, the GFRP deck material never entered into its non-linear range of material behavior. Truck positioning wa s consistent and repeatable. These load strain curves were generated from the static load test data (rolling load cases), and were envisaged to be used as a conversion from the monitoring strain to the wheel load experienced by the bridge during the monitori ng period. Peak strains obtained from the monitoring will be used to convert into wheel lo ad by using these plots. The monitored strains will have some built in dynamic (or impact) component. 76

PAGE 77

Table 6-1. Maxim um static strain values. Instrumentation One truck in lane 1 (TP2) One truck in lane 2 (TP5) Two trucks (TP2+TP5) S1 47.95 0.28 26.69 S2 5.38 -0.44 -3.03 S4 -21.22 20.74 -5.57 S5 84.28 -4.80 58.12 S6 106.82 -16.40 184.23 S7 -6.35 442.29 417.75 B1 147.54 9.37 148.21 B2 118.73 41.75 160.20 B3 66.45 81.34 146.14 B4 8.03 102.73 105.40 Table 6-2. Load di s tribution factors. Strain gage Distribution Factor Strain rosette Distribution Factor S1 0.26 R1 0.33 S2 0.20 R2 0.35 S5 0.27 R5 0.37 S7 0.22 R7 0.46 77

PAGE 78

Time (Sec)Microstrain 0 20 40 60 80 100 120 -100 0 100 200 300 400 500 600 700 800 S7 Figure 6-1. Typical strain-time history. 78

PAGE 79

A B C D Figure 6-2. Influence lines for positive bending for A) S1 for TP1 B) S2 for TP1 C) S5 for TP1 D) S7 for TP5. 79

PAGE 80

A B Figure 6-3. Influence lines for negative be nding for A) S4 for TP5 B) S6 for TP1. 80

PAGE 81

A B Figure 6-4. Distance between ax les A) Measured B) Actual. 81

PAGE 82

Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -800 -600 -400 -200 0 200 400 600 800 TP1 9.5 kip 13 kip 15.5 kip 18 kip AWheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -800 -600 -400 -200 0 200 400 600 800 TP1 9.5 kip 13 kip 15.5 kip 18 kipB Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -800 -600 -400 -200 0 200 400 600 800 TP1 9.5 kip 13 kip 15.5 kip 18 kip CWheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -800 -600 -400 -200 0 200 400 600 800 TP5 9.5 kip 13 kip 15.5 kip 18 kipD Figure 6-5. Influence lines for rose tte A) R1 B) R2 C) R5 D) R7. 82

PAGE 83

Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -100 0 100 200 300 400 500 600 700 800 S5 S6 Figure 6-6. Influence lines fo r gage S5 and S6 for TP1. Figure 6-7. Relative location of ga ges S5 and S6 and maximum strain. 83

PAGE 84

Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -200 -100 0 100 200 300 400 500 600 700 S4 S7 Figure 6-8. Influence lines for gage S4and S7 for TP5. Figure 6-9. Relative location of ga ges S4 and S7 and maximum strain. 84

PAGE 85

A B Figure 6-10. Partial influence lines for soffit gage S7 A) at axle P4 B) at axle P1. Displacement (in.) Displacement (mm)Wheel Load (Kips) Wheel Load (kN) -0.01 0.01 0.03 0.05 0.07 0.09 0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5 Figure 6-11. Load displacement for the bridge deck. 85

PAGE 86

Time (Sec)Displacement (in.) Displacement (mm) 0 20 40 60 80 100 120 -0.01 0 0.01 0.02 0.03 0.04 0.05 0.06 -0.15 0.15 0.45 0.75 1.05 1.35 D2 D3 ATime (Sec)Displacement (in.) Displacement (mm) 0 20 40 60 80 100 120 -0.01 0 0.01 0.02 0.03 0.04 0.05 0.06 -0.15 0.15 0.45 0.75 1.05 1.35 D2 D3B Figure 6-12. Displacement time history for the st eel girders A) truck in TP1 B) truck in TP5. A Bottom Section Flange E (Lengthwise) = 3600 ksi E (Crosswise) = 1600 ksi Bottom Section Base Plate E (Lengthwise) = 2500 ksi E (Crosswise) = 1300 ksi Bottom Section Web E (Lengthwise) = 2600 ksi E (Crosswise) = 1200 ksi Top Sheet E (Lengthwise) = 1500 ksi E (Crosswise) = 800 ksi B Figure 6-13. GFRP deck Modulus map A) Description B) Differe nt elastic modulus. 86

PAGE 87

A B C D Figure 6-14. N.A. locations for panel B10 A) Maximum strain in S1 and corresponding strain in R1 B) Minimum strain in R1 and correspond ing strain in S1 C) Maximum strain in S2 and corresponding strain in R2 D) Minimu m strain in R2 and corresponding strain in S2. 87

PAGE 88

A B C D Figure 6-15. N. A. locations for panel B9 A) Ma ximum strain in S5 and corresponding strain in R5 B) Minimum strain in R5 and correspond ing strain in S5 C) Maximum strain in S7 and corresponding strain in R7 D) Minimu m strain in R7 and corresponding strain in S7. 88

PAGE 89

Figure 6-16. Location of measured and calculated elastic N.A. 89

PAGE 90

Figure 6-17. Strain in extrem e bottom fiber of steel girder. 90

PAGE 91

Axle location from strain gage (ft)MicrostrainAxle location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 0 50 100 150 200 250 300 TP2 Load at axle P5: 19 kip 26 kip 15.5 kip 36kip AAxle location from strain gage (ft)MicrostrainAxle location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 0 50 100 150 200 250 300 TP3 Load at axle P5: 19 kip 26 kip 15.5 kip 36kipB Axle location from strain gage (ft)MicrostrainAxle location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 0 50 100 150 200 250 300 Load at axle P5: TP4 19 kip 26 kip 15.5 kip 36kip CAxle location from strain gage (ft)MicrostrainAxle location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 0 50 100 150 200 250 300 TP5 Load at axle P5: 19 kip 26 kip 15.5 kip 36kipD Figure 6-18. Strain influence lines for full bridge gage A) B1 B) B2 C) B3 D) B4. 91

PAGE 92

A B Figure 6-19. Test set up for A) La b test (Credit: Vyas et al. 2006) B) Field test (Credit: Abhay P. Singh). A B Figure 6-20. Instrumentation and application of load for lab test A) FAIL 1 B) FAIL 2. 92

PAGE 93

A B Figure 6-21. Lab test A) Load Strain B) Load Displacement relation for failure test. 93

PAGE 94

Figure 6-22. Typical influence line illust rating calculation of distribution factor. 94

PAGE 95

Web location from soffit gage (in)MicrostrainWheel location from soffit gage (m) -150 -100 -50 0 50 100 150 -3 -2 -1 0 1 2 3 0 100 200 300 400 500 600 700 800 P5 18 kip AWeb location from soffit gage (in)MicrostrainWeb location from soffit gage (m) -150 -100 -50 0 50 100 150 -3 -2 -1 0 1 2 3 0 100 200 300 400 500 600 700 800 P4 18 kipB Web location from soffit gage (in)MicrostrainWeb location from soffit gage (m) -150 -100 -50 0 50 100 150 -3 -2 -1 0 1 2 3 0 100 200 300 400 500 600 700 800 P5 18 kip CWeb location from soffit gage (in)MicrostrainWeb location from soffit gage (m) -150 -100 -50 0 50 100 150 -3 -2 -1 0 1 2 3 0 100 200 300 400 500 600 700 800 P4 18 kipD Figure 6-23. Modified influence lines for distribu tion factor calculations A) S1 B) S2 C) S5 D) S7. 95

PAGE 96

Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -3 -2 -1 0 1 23 -0.8 -0.4 0 0.4 0.8 -800 -600 -400 -200 0 200 400 600 800 TP1 9.5 kip 13 kip 15.5 kip 18 kip Figure 6-24. Partial influence lines for web gage R1 at axle P5. 96

PAGE 97

A B Figure 6-25. Strain rosettes reflecting A) Positive shear B) Negative shear. 97

PAGE 98

Wheel location from strain gage (ft)MicrostrainWheel location from strain gage (m) -60 -50 -40 -30 -20 -10 0 10 2030 -16 -12 -8 -4 0 4 8 -700 -500 -300 -100 100 300 500 700 TP1 Soffit gage S1 Web gage R1 Figure 6-26. Comparison of Influence lines for soffit and web gage for 18 kip of wheel load. 98

PAGE 99

Wheel location from web gage (in)MicrostrainWheel location from web gage (mm) -27 -18 -9 0 9 1827 -500 -250 0 250 500 -800 -700 -600 -500 -400 -300 -200 -100 0 100 TP1 P5 AWheel location from web gage (in)MicrostrainWheel location from web gage (mm) -27 -18 -9 0 9 1827 -500 -250 0 250 500 -800 -700 -600 -500 -400 -300 -200 -100 0 100 TP1 P5B Wheel location from web gage (in)MicrostrainWheel location from web gage (mm) -27 -18 -9 0 9 1827 -500 -250 0 250 500 -800 -700 -600 -500 -400 -300 -200 -100 0 100 P5 TP1 CWheel location from web gage (in)MicrostrainWheel location from web gage (mm) -27 -18 -9 0 9 1827 -500 -250 0 250 500 -800 -700 -600 -500 -400 -300 -200 -100 0 100 TP5 P4D Figure 6-27. Modified influence lines for distribu tion factor calculations A) S1 B) S2 C) S5 D) S7. 99

PAGE 100

MircostrainWheel Load (Kips) Wheel Load (kN) 0 100 200 300 400 500 600 700800 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5 AMicrostationWheel load (kips) Wheel load (kN) 0 100 200 300 400 500 600 700800 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5B MircostrainWheel Load (Kips) Wheel Load (kN) 0 100 200 300 400 500 600 700800 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5 CMircostrainWheel Load (Kips) Wheel Load (kN) 0 100 200 300 400 500 600 700800 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5D Figure 6-28. Load Strain Calibration Curve for positive bending gage A) S1 B) S2 C) S5 D) S7. MicrostrainWheel Load (Kips) Wheel Load (kN) -210 -180 -150 -120 -90 -60 -30 030 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5AMicrostrainWheel Load (Kips) Wheel Load (kN) -105 -75 -45 -15 15 45 0 5 10 15 20 0 20 40 60 80 TP1 TP2 TP3 TP4 TP5B Figure 6-29. Load Strain Calibration Curve for negative bending gage A) S4 B) S6. 100

PAGE 101

Stress (ksi)Axle Loads ( P4 + P5 ) kip Axle Loads ( P4 + P5 ) kN 0 2 4 6 810 0 20 40 60 80 0 80 160 240 320 TP1 TP2 TP3 TP4 TP5 AStress (ksi)Axle Loads ( P4 + P5 ) kip Axle Loads ( P4 + P5 ) kN 0 2 4 6 810 0 20 40 60 80 0 80 160 240 320 TP1 TP2 TP3 TP4 TP5B Stress (ksi)Axle Loads ( P4 + P5 ) kip Axle Loads ( P4 + P5 ) kN 0 2 4 6 810 0 20 40 60 80 0 80 160 240 320 TP1 TP2 TP3 TP4 TP5 CStress (ksi)Axle Loads ( P4 + P5 ) kip Axle Loads ( P4 + P5 ) kN 0 2 4 6 810 0 20 40 60 80 0 80 160 240 320 TP1 TP2 TP3 TP4 TP5D Figure 6-30. Load Stress plots for full br idge gages A) B1 B) B2 C) B3 D) B4. 101

PAGE 102

CHAP TER 7 SUMMARY AND CONCLUSION This thesis described a bridge load test conducted on a newly installed GFRP composite deck in Belle Glade, Florida. The GFRP comp osite deck replaced a deteriorating steel grid bridge deck, which had deteriorat ed. The deck was supported on st ructural steel girders. The bridge was instrumented using strain gages a nd displacement gages on both the GFRP deck and steel superstructure. A GPS system was used to monitor the position of th e truck while readings were taken. During the bridge load test, strain s and deflections were monitored for linearity to ensure that the bridge was not da maged during the test. The load test results will be used to classify the load level experienced by the bridge during th e monitoring period. Measured GFRP strains were found to be quite localized due to the system flexibility. Peak values of strain from the test truck matched well with the strains measured in a laboratory test. The Following conclusions were ma de based on the bridge test results. Reference strains and deflections Reference strains and deflection were measured These initial strains and deflection will serve the purpose of the base li ne for the long-term monitoring. Maximum GFRP deck strain Maximum strain measured in GFRP deck for the wheel load of 18 kips was 750 This strain was measured at the midspan of the GFRP deck panel. Maximum strain in steel girder Maximum strain measured in the ex isting steel superstructure was 280 measured at mid-span of the bridge. Lateral load distribution Average distribution factor was 0.24 based on the flexural strain and 0.38 based on the shear strain. Strain measurements dropped to 27% of the peak stra in reading when the wheel had moved only 16 in. away from the stra in gage. Measured st rain drops to 10% of its peak value merely by crossing one panel width. Maximum deck deflection Maximum relative deflection measured during the bridge test was 0.09 in., which is 50% more than the service limit deflection (span/800 = 0.06 in.) recommende d by AASHTO LRFD. Deflection recommendation provided by AASHTO was developed for steel and 102

PAGE 103

reinforced concrete bridges. GFRP decks ar e more flexib le compare to steel and RC decks and expected to undergo significant deflecti on under the same wheel load. Maximum relative deflection of steel girders is almost half (55%) of the maximum GFRP deflection recorded for 18 kip of wheel load. Composite action No composite action exists between top GFRP pl ate and bottom FRP panel. It is, therefore concluded that the top plate did not contribute to the flexural stiffness of the bridge deck. Factor of safety The maximum stress calculated based on the meas ured strain in the steel girders was 8.2 ksi. Existing steel superstructure was made of 36 ksi steel. The factor of safety was therefore observed to be 4.4. 103

PAGE 104

APPENDIX DETAILE D LOAD TEST PROCEDURE Detailed load test procedure and existing fram ing plan were presented in this appendix. Took zero (initial) reading and started AE. Following procedure followed throughout the load test. LC 1 (12 Blocks): Truck was rolled slowly in lane 1 at position 2. This position was devised to make truck in the middle of lane 1. Adjusted truck location to maximize strain and/or deflection, truck was stopped at th is location and static data was recorded. Left truck in place, marked this position on deck LC 2 (12 Blocks): Keeping truck in lane 1, slowly rolled second truck in lane 2 at position 5. This position was devised to make truck in th e middle of lane 2. Adjusted truck location to maximize strain and/or deflections, stopped the truc k at this location. Recorded the static data, left truck in place. Marked this position on deck LC 3 (12 Blocks): Removed truck from lane 1, took static reading. Removed truck from lane 2. After LC1 through LC3 were completed, the following steps were followed: 1. Zero reading and restarted AE 2. Rolled 12 block truck in each of th e five positions (LC4 through LC8) 3. Zero reading and restarted AE 4. Rolled 18 block truck in each of th e five positions (LC9 through LC13) 5. Zero reading and restarted AE 6. Rolled 24 block truck in each of th e five positions (LC14 through LC18) 7. Zero reading and restarted AE 104

PAGE 105

105 8. Rolled 30 block truck in each of th e five positions (LC19 through LC23) 9. Terminated DAQ and AE collection. End of the load test.

PAGE 106

LIST OF REFERE NCES Alampalli S., Jonathan, K. (2002)., Rehabilitation and field testing of an FRP bridge deck on a truss bridge, Compos. Struct., 57(1-4), 373-375. American Association of State Highway a nd Transportation Offi cials(AASHTO). (2007). LRFD standard specifications for highway bridges, 4th Edition with 2007 interim revisions., Washington D.C. Park, K.T., Jeong, S.M., Kim, S.J. and Hwang Y.K. (2004), Reliability analysis on GFRP deck for design criteria suggestion, KSCE J. of Civil Eng., 8(3), 301-305. Plunkett, J.D. (1997) Fiber-r einforced polymer honeycomb s hort span bridge for rapid installation. Idea Project Final Rep., Contract NCHRP 96 IDO30, Transportation Research Board, Washington D.C. Vyas, J. S., Zhao, L., Ansely, M.A. and Xia, J ., and LoSciuto, J. (2007), Characterization of a Low-Profile Fiber-Reinforced Polymer D eck System for Moveable Bridges, ASCE J. Bridge Eng., 14(1).55. 106

PAGE 107

BIOGR APHICAL SKETCH The author was born at Rath, Uttar Pradesh (U.P.), in India. Author completed his bachelors degree in Civil Engi neering in May, 2005 from the Indian Institute of Technology (IIT), Roorkee, India. After completing his undergraduate studies, he worked for three years for a bridge design firm in India. The author enrolled in graduate school at the University of Florida in August 2008, where he anticipates receiving the degree of Master of Engineering. The author plans to continue work toward a Doctor of Ph ilosophy degree at the Un iversity of Florida upon graduation. 107