<%BANNER%>

Evaluation of Dual Cable Signal Support Systems

Permanent Link: http://ufdc.ufl.edu/UFE0043803/00001

Material Information

Title: Evaluation of Dual Cable Signal Support Systems
Physical Description: 1 online resource (173 p.)
Language: english
Creator: Rigdon, Jessica L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: cable -- catenary -- hanger -- hurricane -- light -- load -- messenger -- signal -- structure -- support -- traffic -- wind
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: The primary objective of this project was to identify which dual cable support systems provide the best potential for hurricane resistance. Along with hurricane resistance, the serviceability limit for each system was considered. Systems were evaluated using full-scale tests with the UF Hurricane Simulator. Thirty-three wind load tests were performed to measure signal rotations, cable tensions, and cable displacements. The data was analyzed to determine the effect of signal orientation, wind angle, hanger type, and signal material on system performance during high wind conditions. The results were used to calculate force coefficients for dual cable systems. The force coefficients may be applied to wind loads on traffic signals when designing intersections. Results showed that there is a significant difference in rotations and cable tension among the different support systems. The systems that tend to have greater rotation under relatively low wind loads also reduce the increase in cable tension experienced under high wind loads.
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 Jessica L Rigdon.
Thesis: Thesis (M.E.)--University of Florida, 2011.
Local: Adviser: Cook, Ronald A.
Local: Co-adviser: Masters, Forrest.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043803:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043803/00001

Material Information

Title: Evaluation of Dual Cable Signal Support Systems
Physical Description: 1 online resource (173 p.)
Language: english
Creator: Rigdon, Jessica L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: cable -- catenary -- hanger -- hurricane -- light -- load -- messenger -- signal -- structure -- support -- traffic -- wind
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: The primary objective of this project was to identify which dual cable support systems provide the best potential for hurricane resistance. Along with hurricane resistance, the serviceability limit for each system was considered. Systems were evaluated using full-scale tests with the UF Hurricane Simulator. Thirty-three wind load tests were performed to measure signal rotations, cable tensions, and cable displacements. The data was analyzed to determine the effect of signal orientation, wind angle, hanger type, and signal material on system performance during high wind conditions. The results were used to calculate force coefficients for dual cable systems. The force coefficients may be applied to wind loads on traffic signals when designing intersections. Results showed that there is a significant difference in rotations and cable tension among the different support systems. The systems that tend to have greater rotation under relatively low wind loads also reduce the increase in cable tension experienced under high wind loads.
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 Jessica L Rigdon.
Thesis: Thesis (M.E.)--University of Florida, 2011.
Local: Adviser: Cook, Ronald A.
Local: Co-adviser: Masters, Forrest.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043803:00001


This item has the following downloads:


Full Text

PAGE 1

1 EVALUATION OF DUAL CABLE SIGNAL SUPPORT SYSTEMS By JESSICA L. RIGDON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE MASTER OF ENGINEERING UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Jessica L. Rigdon

PAGE 3

3 ACKNOWLEDGMENTS The authors would like to thank the Florida Department of Transportation (FDOT) for providing the funds for this project. The FDOT State Traffic Engineering and Operations Offic e assisted in planning for numerous full scale tests and organized the donation of necessary equipment from various traffic hardware manufacturers. We thank the manufacturers that sent the donated equipment and took the time to attend test days. The author s also thank the City of Gainesville Traffic Operations for their unlimited support by providing experienced teams for the installation of traffic signals and changing equipment between tests. Their hard work and patience was greatly appreciated. Finally, thanks are given to all of our fellow researchers at the Powell Hurricane Research Laboratory who gave their time to this project to ensure the Hurricane Simulator and instrumentation worked smoothly.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 7 LIST O F ABBREVIATIONS ................................ ................................ ........................... 16 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 18 2 LITER ATURE REVIEW ................................ ................................ .......................... 21 3 METHODOLOGY ................................ ................................ ................................ ... 32 3.1 Test Setup ................................ ................................ ................................ ...... 32 3.2 Instrumen tation ................................ ................................ .............................. 34 3.3 Test Method ................................ ................................ ................................ ... 35 4 TEST RESULTS ................................ ................................ ................................ ..... 47 4.1 Signal Rotat ions ................................ ................................ ............................. 47 4.1.1 Effect of Signal and Simulator Orientations on Signal Rotation ........... 47 4.1.2 Effect of Hanger Types and Support Systems on Signal Rotation ....... 48 4.1.3 Effect of Multiple Signals on Signal Rotation ................................ ....... 51 4.1.4 Effect of Signal Weight on Signal Rotation ................................ .......... 52 4.1.5 Effect of Messenger Cable Clamp Location on Signal Rotation .......... 53 4.2 Cable Tensions ................................ ................................ .............................. 53 4.2.1 Effect of Signal and Simulator Orientations on Cable Tension ............ 54 4.2.2 Effect of Hanger Types and Support Systems on Cable Tension ........ 55 4.2.3 Effect of Multiple Signals on Cable Tension ................................ ........ 56 4.2.4 Effect of Signal Material on Cable Tension ................................ .......... 57 4.2.5 Effect of Messenger Clamp Location on Cable Tension ...................... 57 4.3 Cable Displacements ................................ ................................ ..................... 58 5 FORCE COEFFICIENTS F OR WIND FORCES ................................ ................... 106 6 PERFORMANCE OF EQUIPMENT AND HARDWARE ................................ ....... 110 7 CONCLUSIONS, SUMMARY, AND RECOMMENDATIONS ................................ 118 7.1 Summary ................................ ................................ ................................ ...... 118

PAGE 5

5 7.2 Conclusions ................................ ................................ ................................ .. 118 7.3 Recommendations ................................ ................................ ....................... 120 APPENDIX A SIGNAL ROTATION VS. WIND VELOCITY GRAPHS ................................ ......... 123 B CABLE TENSIONS VS. WIND VELOCITY GRAPHS ................................ ........... 139 C CABLE DISPLACEMENT VS. WIND VELOCITY GRAPHS ................................ 155 D EQUIPMENT CATALOG ................................ ................................ ...................... 167 LIST OF REFERENCES ................................ ................................ ............................. 171 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 173

PAGE 6

6 LIST OF TABLES Table page 1 1 Traffic signal statistics for 2004 hurricane season (FDOT 2005) ........................ 20 3 1 Test Schedule ................................ ................................ ................................ ..... 46 4 1 Changes in cable tension for Test Series 2 ................................ ...................... 103 4 2 Changes in cable tension for Test Series 3 ................................ ...................... 103 4 3 Changes in cable tension for Test Series 4 ................................ ...................... 104 4 4 Changes in cable tension for Test Series 5 ................................ ...................... 104 4 5 Changes in cable tension for Test Series 6 ................................ ...................... 105

PAGE 7

7 LIST OF FIGURES Figure page 2 1 Signals supported by dual cable s failed during Hurricane Andrew ..................... 27 2 2 Three test configurations from Cook and Johnson (2007) ................................ .. 28 2 3 Configuration of signal and cables ................................ ................................ ...... 28 2 4 Limiting rotation for functional visibility of incandescent light signals (Cook 1993) ................................ ................................ ................................ .................. 29 2 5 Vert when using LED signals (sideways view of horizontal rotation) ................................ ................................ ................................ .............. 29 2 6 Horiz for visibility of LED signals (top view of vertical rotation) ................................ ................................ ................................ .............. 30 2 7 Free body diagram of signal supported by single cable system (Cook and Johnson 2007) ................................ ................................ ................................ .... 31 3 1 Spacing and layout of cables ................................ ................................ .............. 37 3 2 Cable support systems ................................ ................................ ....................... 38 3 3 nes and fans (back) ................................ .... 39 3 4 ............................... 39 3 5 Signal and simulator testing orient ations ................................ ............................ 40 3 6 Two signal test setup ................................ ................................ ......................... 40 3 7 R.M. Young anemometer ................................ ................................ .................. 41 3 8 Load cells installed in line with catenary and messenger cables ....................... 41 3 9 String potentiometer setup ................................ ................................ ................. 42 3 10 Mounting sy stem for string potentiometers ................................ ........................ 42 3 11 Orientation sensor installed inside the signal ................................ .................... 43 3 12 Types of hangers ................................ ................................ ............................... 44 3 13 Sample of oscillating wind load sequence (Test 4) ................................ ............. 44 3 14 Sample of linear wind load (Test 1) ................................ ................................ .... 45

PAGE 8

8 4 1 Rotation for single cable at all orientations (Tests 7, 15, 16, and 33) ................. 60 4 2 Rotation for pipe hanger at all orientations (Tests 4, 9, 10, and 28) ................... 61 4 3 Rotation for pivotal hanger at all orientations (Tests 5, 11, 12, and 29) .............. 61 4 4 Rotation for cable hanger at all orientations (Tests 6, 13, 14, and 30) ............... 62 4 5 Rotation for systems at 90 signal and 90 simulator orientation (Test Series 2) ................................ ................................ ................................ ........................ 63 4 6 Rotation for systems at 90 signal and 45 simulator orientations (Tests 9, 11, 13, 15, and 23) ................................ ................................ ............................. 64 4 7 Adjustable strap hanger with visible yielding in lower section (Test 23) ............ 64 4 8 Rotation for systems at 45 signal and 45 simulator orientations (Tests 10, 12, 14, and 16) ................................ ................................ ................................ ... 65 4 9 Rotation for systems at 90 signal and 12 simula tor orientations (Tests 28, 29, 30, and 33) ................................ ................................ ................................ ... 66 4 10 Rotation for 5 section signals during Test Series 4 two at 90 signal and 45 simulator orientations ................................ ................................ ......................... 67 4 11 Rotation for 3 section signals during Test Series 4 two at 90 signal and 45 simulator orientations ................................ ................................ ......................... 68 4 12 Rotation for single cable system with two signals verses one signal (Tests 15 and 21) ................................ ................................ ................................ ............... 69 4 13 Rotation for pipe hanger with two signals verses one signal (Tests 9 and 17) ... 70 4 14 Rotation for pivotal hanger with two signals verses one signal (Tests 11 and 18) ................................ ................................ ................................ ...................... 71 4 15 Rotation for cable hanger with two signals verses one signal (Tests 13 and 19) ................................ ................................ ................................ ...................... 72 4 16 Rotation for single cable system with aluminum signal verses polycarbonate signal at 90 signal and 90 simulator orientation (Tests 7 and 8) ...................... 7 3 4 17 Rotation for single cable system with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 15 and 22) ............................ 74 4 18 Rotation for single cable system with al uminum verses polycarbonate signal at 90 signal and 12 simulator orientation (Tests 33 and 32) ............................ 75

PAGE 9

9 4 19 Rotation for pivotal hanger with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 11 and 25) ................................ ....... 76 4 20 Rotation for pivotal hanger with aluminum verses polycarbonate signal at 90 signal and 12 simulator orientation (Tests 29 and 31) ................................ ....... 77 4 21 Rotation for cable hanger with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 13 and 26) ................................ ....... 78 4 22 Rotation for pipe hanger with altered messenger clamp location (Tests 9 and 27) ................................ ................................ ................................ ...................... 79 4 23 Rotation for pivotal hanger with altered messenger clamp location (Tests 11 and 24) ................................ ................................ ................................ ............... 80 4 24 Change in tension for single cable at all orientations (Tests 7, 15, 16, and 33) .. 81 4 25 Change in tension for pipe hanger at all orientations (Tests 4, 9, 10, and 28) .... 82 4 26 Change in tension for pivotal hanger at all orientations (Tests 5, 11, 12, and 29) ................................ ................................ ................................ ...................... 83 4 27 Change in tension for cable hanger at all orientations (Tests 6, 13, 14, and 30) ................................ ................................ ................................ ...................... 84 4 28 Change in tension for all systems at 90 signal and 90 simulator orientations .. 85 4 29 Change in tension for all systems at 90 signal and 45 simulator orientations .. 86 4 30 Change in tension for all systems at 45 signal and 45 simulator orientations .. 87 4 31 Change in tension for all systems at 90 signal and 12 simulator orientations .. 88 4 32 C hange in tension for all systems with two signals ................................ ............. 89 4 33 Change in tension for single cable system with two verses one signal (Tests 15 and 21) ................................ ................................ ................................ .......... 90 4 34 Change in tension for pipe hanger with two verses one signal (Tests 9 and 17) ................................ ................................ ................................ ...................... 91 4 35 Change in tension for pivotal hanger with two verses one signal (Tests 11 and 18) ................................ ................................ ................................ ............... 92 4 36 Change in tension for cable hanger with two signals verses one signal (Tests 13 and 19) ................................ ................................ ................................ .......... 93

PAGE 10

10 4 37 Change in tension for t he single cable system with aluminum verses polycarbonate signal at 90 signal and 90 simulator orientations (Tests 7 and 8) ................................ ................................ ................................ ................. 94 4 38 Change in tension for the single cable system with aluminu m verses polycarbonate signal at 90 signal and 45 simulator orientations (Tests 15 and 22) ................................ ................................ ................................ ............... 95 4 39 Change in tension for the single cable system with aluminum verses polycarbonate signal at 90 signal and 12 simulator orientations (Tests 33 and 32) ................................ ................................ ................................ ............... 96 4 40 Change in tension for pipe hanger with altered messenger clamp location (Tests 9 and 27) ................................ ................................ ................................ 97 4 41 Change in tension for pivotal hanger with altered messenger clamp location (Tests 11 and 24) ................................ ................................ ............................... 98 4 42 Cable displacement for pipe hanger at 90 signal and 90 simulator orientation ................................ ................................ ................................ ........... 99 4 43 Cable displacement for pivotal hanger at 90 signal and 90 simulator orientation ................................ ................................ ................................ ......... 100 4 44 Ca ble displacement for cable hanger at 90 signal and 90 simulator orientation ................................ ................................ ................................ ......... 101 4 45 Cable displacement for single cable system at 90 signal and 90 simulator orientation ................................ ................................ ................................ ......... 102 5 1 Free body diagrams for traffic signals ................................ ............................... 108 5 2 Top view of wind force applied to signal with reactions in messenger cable .... 108 5 3 Force coefficient vs. signal rotation for dual cable systems .............................. 109 5 4 Force coefficient vs. wind velocity for dual cable systems ................................ 109 6 1 Failure of the joint between the top signal head and two way bracket (Test 22) ................................ ................................ ................................ ................... 114 6 2 Worn serrated teeth on two way bracket which connects to top signal head (Test 22) ................................ ................................ ................................ .......... 114 6 3 Failure at connection for the backplate on polycarbonate signal (Test 22) ...... 115 6 4 Failed backplate material at connection point (Test 27) ................................ .. 115 6 5 Failed sand cast aluminum stabilizer clamp ................................ ..................... 116

PAGE 11

11 6 6 Spa n wire clamp with broken piece ................................ ................................ 117 A 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) (Hanger failure at 115 mph) ................................ ................................ .... 123 A 2 Test 2: Pivotal Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) ................................ ................................ ................................ ................ 123 A 3 Test 3: Direct Connection with 90 Signal and 90 Simulator Orientation s (Linear Load) ................................ ................................ ................................ .... 124 A 4 Test 4: Pipe Hanger with 90 Signal and 90 Simulator Orientations ............... 124 A 5 Test 5: Pivotal Hanger with 9 0 Signal and 90 Simulator Orientations ............ 125 A 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator Orientations ............. 125 A 7 Tes t 7: Direct Connection with 90 Signal and 90 Simulator Orientations ....... 126 A 8 Test 8: Direct Connection with 90 Signal and 90 Simulator Orientations (Polycarbonate Signal) (Backplate failure at 107 mph) ................................ ..... 126 A 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations ............... 127 A 10 Test 10: Pipe Hanger with 45 S ignal and 45 Simulator Orientations ............. 127 A 11 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations .......... 127 A 12 Te st 12: Pivotal Hanger with 45 Signal and 45 Simulator Orientations .......... 128 A 13 Test 13: Cable Hanger with 90 Signal and 45 Simulator Orientations ........... 128 A 14 Test 14: Cable Hanger with 45 Signal and 45 Simulator Orientations ........... 129 A 15 Test 15: Direct Connection with 90 Signal and 45 Simulator Orientations ..... 129 A 16 Test 16: Direct Connection with 45 Signal and 45 Simulator Orientations ..... 129 A 17 Test 17: Pipe Hanger with 90 Signal and 45 Sim ulator Orientations (Two Signals) ................................ ................................ ................................ ............ 130 A 18 Test 18: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 130 A 19 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 131 A 20 Test 20: Discontinuous Messenger with 90 Signal and 45 Simulator Orientations (Two Signal s) ................................ ................................ ............... 131

PAGE 12

12 A 21 Test 21: Direct Connection with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ .... 132 A 22 Test 22: Direct Co nnection with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) (Signal Head Failure at 118 mph) ................................ 132 A 23 Test 23: Adjustable Strap Hanger with 90 Signal and 45 Simulator Orie ntations ................................ ................................ ................................ ...... 133 A 24 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front) ................................ ................................ ....................... 133 A 25 Test 25: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) ................................ ................................ ..................... 134 A 26 Test 26: Cable Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) ................................ ................................ ..................... 134 A 27 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front) (Backplate failure at 115 mph) ................................ ...... 135 A 28 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations ............. 135 A 29 Test 29: Pivotal Hanger with 90 Signal and 12 Simulator Orientations .......... 136 A 30 Test 30: Cable Hanger with 90 Signal and 12 Simulator Orientations ........... 136 A 31 Test 31: Pivotal Hanger with 90 Signal and 12 Simulator Orientations (Polycarbo nate Signals) ................................ ................................ ................... 137 A 32 Test 32: Direct Connection with 90 Signal and 12 Simulator Orientations (Polycarbonate Signals) ................................ ................................ ................... 137 A 3 3 Test 33: Direct Connection with 90 Signal and 12 Simulator Orientations ..... 138 B 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) (Hanger failure at 115 mph) ................................ ................................ .... 139 B 2 Test 2: Pivotal Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) ................................ ................................ ................................ ................ 139 B 3 Test 3: Direct Connection w ith 90 Signal and 90 Simulator Orientations (Linear Load) ................................ ................................ ................................ .... 140 B 4 Test 4: Pipe Hanger with 90 Signal and 90 Simulator Orientations ............... 140 B 5 Test 5: Pivotal Hanger with 90 Signal and 90 Simulator Orientations ............ 141 B 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator Orientations ............. 141

PAGE 13

13 B 7 Test 7: Direct Connection with 90 Signal and 90 Simulator Orientations ....... 141 B 8 Test 8: Direct Connection with 90 Signal and 90 Simulator Orientatio ns (Polycarbonate Signal) (Backplate failure at 107 mph) ................................ ..... 142 B 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations ............... 142 B 10 Test 10: Pipe Hanger with 45 Signal and 45 Simulator Orientations ............. 143 B 11 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations .......... 143 B 12 Test 12: Pivotal Hanger with 45 Signal and 45 Simulator Orientations .......... 143 B 13 Test 13: Cable Hanger with 90 Signal and 45 Simulator Orientations ........... 144 B 14 Test 14: Cable Hanger with 45 Signal and 45 Simulator Orientations .......... 144 B 15 Test 15: Direct Connection with 90 Signa l and 45 Simulator Orientations ..... 145 B 16 Test 16: Direct Connection with 45 Signal and 45 Simulator Orientations ..... 145 B 17 Tes t 17: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 146 B 18 Test 18: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 146 B 19 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 147 B 20 Test 20: Discontinuous Messenger with 90 Sign al and 45 Simulator Orientations (Two Signals) ................................ ................................ ............... 147 B 21 Test 21: Direct Connection with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ .... 148 B 22 Test 22: Direct Connection with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) (Signal Head Failure at 118 mph) ................................ 14 8 B 23 Test 23: Adjustable Strap H anger with 90 Signal and 45 Simulator Orientations ................................ ................................ ................................ ...... 149 B 24 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front) ................................ ................................ ....................... 149 B 25 Test 25: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) ................................ ................................ ..................... 150

PAGE 14

14 B 26 Test 26: Cable Hanger with 90 Signal and 45 Simulato r Orientations (Polycarbonate Signal) ................................ ................................ ..................... 150 B 27 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front) (Backplate failure at 115 mph) ................................ ...... 151 B 28 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations ............. 151 B 29 Test 29: Pivotal Hanger with 90 Signal and 12 Simulator Orientati ons .......... 152 B 30 Test 30: Cable Hanger with 90 Signal and 12 Simulator Orientations ........... 152 B 31 Test 31: Pivotal Hanger with 90 Sign al and 12 Simulator Orientations (Polycarbonate Signals) ................................ ................................ ................... 153 B 32 Test 32: Direct Connection with 90 Signal and 12 Simulator Orientations (Polycarbonate Signals) ................................ ................................ ................... 153 B 33 Test 33: Direct Connection with 90 Signal and 12 Simulator Orientations ..... 154 C 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientatio ns (Linear Load)(Hanger failure at 115 mph) ................................ ................................ ..... 155 C 2 Test 2: Pivotal Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) ................................ ................................ ................................ ................ 155 C 3 Test 3: Direct Connection with 90 Signal and 90 Simulator Orientations (Linear Load) ................................ ................................ ................................ .... 156 C 4 Test 4: Pipe Hanger with 90 Signal and 90 Simulator Orientations ............... 156 C 5 Test 5: Pivotal Hanger with 90 Signal and 90 Simulator Orientations ............ 157 C 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator O rientations ............. 157 C 7 Test 7: Direct Connection with 90 Signal and 90 Simulator Orientations ....... 158 C 8 Test 8: Direct Connection w ith 90 Signal and 90 Simulator Orientations (Polycarbonate Signal) (Backplate failure at 107 mph) ................................ ..... 158 C 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations ............... 158 C 10 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations .......... 159 C 11 Test 12: Pivotal Hanger with 45 Signal and 45 Simulato r Orientations .......... 159 C 12 Test 14: Cable Hanger with 45 Signal and 45 Simulator Orientations ........... 160 C 13 Test 15: Direct Connection with 90 Signal and 45 Simulator Orientations ..... 160

PAGE 15

15 C 14 Test 16: Direct Connection with 45 Signal and 45 Simulator Orientations ..... 161 C 15 Test 17: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 161 C 16 Test 18: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 162 C 17 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) ................................ ................................ ................................ ............ 162 C 18 Test 23: Adjustable Strap Hange r with 90 Signal and 45 Simulator Orientations ................................ ................................ ................................ ...... 163 C 19 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger clamped to front of hanger) ................................ ........................... 163 C 20 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Messenger clamped to front of hanger) (Failure of Backplate at 115 mph) ..... 164 C 21 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations ............. 164 C 22 Test 30: Cable Hanger with 90 Signal and 12 Simulator Orientations ........... 165 C 23 Test 31: Pivotal Hanger with 90 Signal and 12 Simulator Orientations (Polycarbonate Signal) ................................ ................................ ..................... 165 C 24 Test 33: Direct Connection with 90 Signal and 12 Simu lator Orientations ..... 166

PAGE 16

16 LIST OF ABBREVIATION S AASHTO American Association of State Highway and Transportation Officials ASCE American Society of Civil Engineers ATLAS Analysis of Traffic Lights and Signs direct connect. di rect connection disc. mess. discontinuous messenger FDOT Florida Department of Transportation ITE Institute of Transportation Engineers lbs force pounds mph miles per hour string p ots string p otentiometers poly. polycarbonate UF University of Florida

PAGE 17

17 Abs tract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Engineering EVALUATION OF DUAL CABLE SIGNAL SUPPORT SYSTEMS By Jessica L. Rigdon December 2011 Cha ir: Ronald A. Cook Cochair: Forrest Masters Major: Civil Engineering The primary objective of this project was to identify which dual cable support systems provide the best potential for hurricane resistance. Along with hurricane resistance, the serviceab ility limit for each system was considered. Systems were evaluated using full scale tests with the UF Hurricane Simulator. Thirty three wind load tests were performed to measure signal rotations, cable tensions, and cable displacements. The data was analyz ed to determine the effect of signal orientation, wind angle, hanger type, and signal material on system performance during high wind conditions. The results were used to calculate force coefficients for dual cable systems. The force coefficients may be ap plied to wind loads on traffic signals when designing intersections. Results showed that there is a significant difference in rotations and cable tension among the different support systems. The systems that tend to have greater rotation under relatively l ow wind loads also reduce the increase in cable tension experienced under high wind loads.

PAGE 18

18 CHAPTER 1 INTRODUCTION The performance of dual cable traffic signal support systems during hurricanes has indicated the need to develop vertical signal hangers and disconnect boxes that have an improved resis tance to hurricane wind loads. During Hurricane Andrew in 199 2 there was severe damage to cable supported traffic signals in Florida. Although the high failure rate encouraged the start of research into the dev elopment of hurricane resistant traffic signal equipment, there was still a high rate of failure of cable supported signals during the hurricane season of 2004. An evaluation of damage to traffic signals following the hurricane season of 2004 showed that additional research and development was necessary for improving the performance of traffic signal equipment (FDOT 2005). The report included the number of signals in the state and the number of damaged systems during that season ( Table 1 1 ability to resist hurricane conditions, it was also noted that damage to cable supported systems was repaired much quicker and easier than damage to mast arms. After reviewing the of installing traffic signals using span wires, strain poles, and hanger devices was Signal f ailures commonly occur in the vertical hanger nearest to the messenger cable attachment or in the connections at either the top or bottom of the disconnect bo x.

PAGE 19

19 These failures are likely due to the high moment that exists in the hanger and disconnect box a t the connection to the messenger cable. Alternatives to the standard dual cable system with pipe hangers or adjustable strap hangers are under consideration for improved hurricane resistan ce of cable supported signals. The alternatives include a pivotal hanger assembly, a cable hanger, and a discont inuous messenger cable system. Previous research by the University of Florida and the Florida Department of Transportation Structures and Traffic Operations Offices compared the standard dual cable hanger suppo rt systems with rigid hangers and a single cable s ystem (Cook and Johnson 2007). The results of this research indicated that a single cable support system would be significantly more hurricane resistant than the current hanger used in the dual cable suppor t system. The results from that research also showed that the single cable system had a lower limit for visibility. The subject of this research was evaluation of the alternatives mentioned above for their potential to improve hurricane resistance and serv iceability. Full scale hurricane wind tests were performed to evaluate these systems. This report summarizes the data from testing and provides comparison among the results of the systems tested.

PAGE 20

20 Table 1 1 Traffic signal stat istics for 2004 hurricane season ( FDOT 2005 ) District No. Total no. of signals district wide Total mast arm signals district wide Total span wire signals district wide Mast arm structural damage Signalized intersections that sustained damage* 1 1,778 802 976 2 496 2 1,585 537 1,048 0 40 3 987 300 687 2 265 4 3,329 1,180 2,149 14 735 5 2,972 458 2,514 2 1,885 6 2,640 1,848 660 0 0 7 2,151 518 1,633 0 102 Sum 15,442 5,643 9,667 20 3,523 Damage defined as loss of signal due to failure of the span wi re, bracket assembly, mast arm mounting hardware or other components

PAGE 21

21 CHAPTER 2 LITERATURE REVIEW This project is related to previous research topics on dual and single cable support systems for traffic signals that were funded by the Florida Department of Transportation (FDOT) and performed by the University of Florida (UF). The focus of these projects was the resistance of various configurations to high velocity wind events, such as hurricanes. Following hurricanes such as Hurricane Andrew, there were a high number of damaged traffic signals. Intersections with damaged signals primarily consisted of dual cable supported systems ( Figure 2 1 ). The most recent project tested and compared dual cable and single cable systems with various cable sag, hangers, weights, and signal orientations to learn more about the forces on the signals, cables, and poles and the signal rotation under high speed wind (Cook and Johnson 2007). An earlier multi phase project developed a com puter program to model cable supported traffic signals under wind loads (Cook et al. 1993; Hoit et al. 1995, 1997). Phase I of the project developed the design standards for cable supported traffic control devices and the Analysis of Traffic Lights and Si gn s (ATLAS) program. During Phase II, full scale wind tests were conducted on cable supported signals and a graphic user interface was incorporated into ATLAS. The final report included changes and enhancements to The ATLA S software is maintained and updated by the Bridge Software Institute at UF. ATLAS predicts the effects of wind loads on various dimensions of signs and signals with differing cable systems, spans, and heights. Currently the program uses rigid hangers, suc h as the pipe and adjustable strap hangers, in its analyses and allows

PAGE 22

22 the user to select dimensions and material of the bracket and hanger; however, flexible or pivoting hangers are not available design options. This project reused the 50 foot span at the Powell Family Structures Laboratory from previous testing. The test configurations used by Cook and Johnson (2007) and the available cable locations are presented in Figure 2 2 The span width was determined to be adequate. An an alysis using ATLAS to compare a 50 foot span to a 72 foot span showed that differences in results were negligible (Cook and Johnson 2007). Referring to the FDOT Manual of Uniform Standards for Design, Construction and Maintenance (2007) confirmed that the 50 foot span represented a realistic intersection width. Traffic signals supported by single cable systems act as pendulums when subjected to wind loads by swinging freely and do not develop stress in the signal hangers or disconnect boxes (Cook and Joh nson 2007). This behavior prevents increased tension in the catenary cable and minimizes hanger and connection failure during extreme wind conditions. When signals are supported by dual cable systems the messenger cable, shown in Figure 2 3 restricts the free swinging movement of the signal and causes high bending moments to build up in rigid hangers. Failure of signal hanger s and connections on dual cable support systems is a frequent problem seen throughout Flori da following hurricanes and other extreme wind events. With traditional rigid hangers used on single and dual cable systems and with wind perpendicular to the cable span, the orientation of a traffic signal had negligible effect on the tension experienced in the catenary cable. Another result of wind loading was that the

PAGE 23

23 messenger cable in dual cable supported traffic signals experienced a severe increase in tension. When a traffic signal experiences rotation due to wind, it eventually reaches a rotation a ngle where functional visibility is limited. The limiting angle for functional visibility was taken as the point at which only half of a bulb is visible to drivers which gnal Figure 2 4 ). These values were calculated with Equation 2 1, which was developed for the by Cook et al. (1993). (2 1) The average wind speeds at which th e 50% visibility was reached for Cook and Johnson in 2007 were 72 miles per hour for dual cable systems and 68 miles per hour for single cable systems ( Figure 2 4 ). The addition of a pivot point on hangers supporti ng traffic signals could potentially limit the signal serviceability, or functional visibility, because of increased rotations. Because this project used light emitting diode (LED) signals, the rotation limits were based on limits for finding the minimum maintained luminous intensity (Institute of Transportation Engineers 2005). The minimum maintained luminous intensity varies with signal color and size and is affected by factors for horizontal and vertical rotations. The allowed ranges of rotations for lu minous intensity calculations were used as the limiting angles for visibility. The vertical rotation limit was 12.5 forward to 27.5 backward in a direction that was perpendicular to the span ( Figure 2 5 ). The hor izontal rotation limit was 27.5 towards either side of the signal ( Figure 2 6 ).

PAGE 24

24 The most recent design standards covering wind loads include the ASCE 7 10 : Minimum Design Loads for Buildings and Other Structures (ASCE 2010), and the Standard Specifications for Structural Supports for Highway Signs, Luminaires and Traffic Signals (AASHTO 2009). These specifications along with information obtained in previous FDOT projects will be taken into consideration to predict wind loads and forces experienced while conducting tests. The ASCE 7 10 provides sufficient information for determining design wind speeds throughout the state of Florida. The ASCE 7 10 uses a wind force coefficient for determining a combined drag and lif t coefficient on structures other than buildings, however there are no appropriate values provided for objects suspended on cables. The values provided are for long objects of a constant cross sectional area for specific cross sectional shapes. During pas t research there has been a concern with determining variable drag and lift coefficients. When traffic signals are placed under wind loads, the rotation that the signals and cables experience causes variability in both drag and lift coefficients. The AASHT O Standard Specifications for Structural Supports for Highway Signs, Luminaires and Traffic Signals provides a constant wind drag coefficient of 1.2 for traffic signals and no consideration for lift (AASHTO 2009). This specification notes that experimental data may be used to modify the drag coefficient based on findings from James F. Marchman, III when a traffic signal is free swinging, but when the swinging of the signal is restrained, as with the dual cable system, the full wind load should be taken on t he signal. When the results from a series of tests from Cook and Johnson were compared higher (Cook and Johnson 2007 ) For the tests by Marchman, the signals were attached to a fixed support, but were allowed to

PAGE 25

25 ro tate below the support. Cook and Johnson used a single cable system permitting rotation of the support as well as the free swinging signal. Cook and Johnson determined that the single cable system behaved like a pendulum in which the support system did not experience increased forces. Cook and Johnson suggested that a drag coefficient of 0.7 and a lift coefficient of 0.4 would be reasonable for single cable systems. When calculating drag and lift coefficients from test data, they used Equation 2 2 and Equat ion 2 3 which coordinate with the free body diagram found in Figure 2 7 from Cook and Johnson 2007. (2 2) (2 3) single cable system. The cable rotation was measured separately from the signal Figure 2 7 ). The resultant tension f Equation 2 4. (2 4) A study in 1997 by the New York State Department of Transportation (NYSDOT) compared AASHTO design procedures from 1983 and 1989 (Alampalli). The design loads were calculated for ap plication to the strain poles of single cable support systems These loads were compared to measured loads from in field testing. Throughout the New York State it was common at the time for engineers to use a spanwire program

PAGE 26

26 that was based on 1983 design procedures. Alampalli found that the 1983 design procedure over designed strain poles when compared to the 1989 design procedures. It was also shown that both AASHTO procedures were conservative when compared to the field tests, especially at higher wind s peeds.

PAGE 27

27 Figure 2 1 Signals supported by dual cables failed during Hurricane Andrew (Photo courtesy of Ronald A. Cook )

PAGE 28

28 Figure 2 2 Three test configurations from Cook and Johnson (2007) Figure 2 3 Configuration of signal and cables

PAGE 29

29 Figure 2 4 Limiting rotation for functional visibility of incandescent light signals (Cook 1993) Figure 2 5 Limiti Vert when using LED signals (sideways view of horizontal rotation)

PAGE 30

30 Figure 2 6 Horiz for visibility of LED signals (top view of vertical rotation)

PAGE 31

31 Figure 2 7 Free body diagram of signal supported by single cable system (Cook and Johnson 2007)

PAGE 32

32 CHAPTER 3 METHODOLOGY 3.1 Test Setup Traffic signals were installed on a 50 foot span with various cable support systems ( Figure 3 1 ). The strain poles used to support the signals were installed during the previous signal related research project at the University of Florida (UF) in 2005. The Gainesville, FL Cook and Johnson used ATLAS to determine that the results for signal rotations and cable displacements were very alike for a 50 foot span and a 72 foot span. The 50 foot span was considered adequate for reuse. tenary and messenger cables was planned in order to maximize the moment developed on the hangers. Two out of the four hangers being tested featured a design that releases moment. After the first test ould be used in order to allow maximum rotations and the most dynamic behavior for signals ( Figure 3 1 ) Backplates are typically used on all signals facing east or west at intersections. All tests used 5 section signals with louv ered backplates to maximize the exposed area of the signal in the wind field. The second signal on tests with two signals was a 3 section signal with a louvered backplate. All test configurations were performed with signals of aluminum, and select tests we re repeated with polycarbonate signals. The weight of the signals and other equipment can be found in Appendix D. Dual cable systems were tested with a 5% sag, and single cable systems were tested with 3% sag ( Figure 3 1 ). A disco ntinuous messenger cable system was also tested using two signals and a 5% sag. In this system, the messenger cable is severed

PAGE 33

33 at the signal on each end of the span so that it is not attached to the strain poles. Figure 3 2 7 wire strand for the span wire as specified in Section 634 of the FDOT 2010 Standard Specifications for Road and Bridge Construction. The City of Gainesville Traffic Operations provided and installed the span wires. Come alongs with a five thousand pound capacity were installed at one end of each cable in order to allow adjustments to cable sag and tension during signal installation. The wind for testing was generate d by the UF Hurricane Simulator ( Figure 3 3 and Figure 3 4 ). The Hurricane Simulator was placed approximately 12 feet away from the signal for each test. During the tests p erformed by Cook and Johnson, wind loads were at the same angle and location for all tests. The Hurricane Simulator was moved during this project to check the sensitivity of the data to the direction of the wind. In Series 1 and 2, the simulator was angled at 90 to the span during testing. In Series 3 to 5, the simulator was placed at 45 to the span. In Series 6, the simulator was angled at 12 to the span. The target of 10 for the final wind angle could not be met because of the size of the simulator an d its proximity to the strain pole and other objects at the test site. The signals were tested facing 90 to the span for all simulator positions, and some tests were repeated with the signal facing 45 to the span for the 45 simulator position. Figure 3 5 shows all of the orientations that were used. The data from Cook and Johnson showed that a forward facing signal had either more rotation or little difference in rotation compared to a backward facing signal. No tests were performed with the signal facing backwards during this project so other variables could be tested.

PAGE 34

34 For tests with one signal on the span, the signal was centered and the simulator was positioned so that the wind field centered on the signal ( Figure 3 5 ). When a second signal was added, the two signals were spaced with five feet between the centers. The two signals were shifted along the span so that the center of the span was between them. The two signal s were situated so that they were completely within the wind field ( Figure 3 6 ). 3.2 Instrumentation The instruments used during testing included an anemometer produced by R.M. Young Company, two LCCA 5K load cells three string potentiometers, and a model 3DM GX2 gyro enhanced orientation sensor. The anemometer was placed approximately six feet in front of the signal in order to measure the velocity of the wind acting on the signal. In order to avoid creating turb ulence directly in front of the signal, the anemometer was shifted two feet to the side of the centerline between the signal and the simulator. The device was mounted on a 1 Figure 3 7 ). One load cell was installed on each of the signal support cables to measure tension of the cables during testing ( Figure 3 8 ). The ends that the load cells were attached to were opposite fro m a pair of come alongs installed with the cables for making adjustments to the cable sag and tension. The load cells had a five thousand pound load capacity, and were S type tension and compression load cells. There were three string potentiometers (stri ng pots) placed in the layout shown in Figure 3 9 The string pots were mounted on a sliding track placed on a tower ten feet behind the signal ( Figure 3 10 ). The slides al lowed adjustments to be made after the signal was installed to ensure that the string pot lines were level. This also allowed for

PAGE 35

35 the string pots to be easily removed for storage when tests were not being performed. The tower was anchored to the ground wit h guy wire lines to keep it from moving in the wind. The 3DM GX2 allowed for the collection of roll, pitch, and yaw measurements to be taken wirelessly. The elimination of wires attached to the signal avoided unnecessary interference with signal movement. The only alteration of the signal was to drill small holes on the internal walls between the heads in order to fasten the sensor. The sensor was mounted inside the casing of the solid yellow signal head ( Figure 3 11 ). 3.3 Test Method The equipment tested consisted of five types of hangers and a direct connection for attaching signals to support cables. A description of each test can be found in Table 3 1 The 2010 FDOT Des cable support systems ( Figure 3 12 A). Other products used in Florida for dual cable support systems include an adjustable strap hanger ( Figure 3 12 B) and a cable hanger ( Figure 3 12 C). A recently developed product, the pivotal hanger, co nsists of the top portion of the adjustable strap hanger paired with a pivoting assembly for the bottom portion which attaches to the messenger cable and disconnect box ( Figure 3 12 D). The adjustable strap hanger w as used when testing the discontinuous messenger system to replicate the practice of a Florida county that developed the system. A single cable support system was tested using a direct connection ( Figure 3 12 E). An oscillating wind load sequence was developed to test the performance of each support system using the Hurricane Simulator. Oscillations simulated turbulence which would be expected during actual hurricanes. The program began with a ramp up to

PAGE 36

36 approximatel y 50 mph over a 30 second period, oscillated through ten cycles at the natural frequency for the system being tested at a mean velocity of 50 mph, ramped to approximately 100 mph over a 30 second period, cycled ten times at the natural frequency at a mean velocity of 75 mph, ramped to 120 mph over a 30 second period, and held that velocity for an additional 30 seconds before the test ended. See Figure 3 13 for a sample of the oscillating wind sequence taken from Tes t 4. The oscillating sequence was used for Series 2 through 6 ( Table 3 1 ). Test Series 1 had a linear wind load to 120 mph over three minutes. Some variation in loading occurred due to natural wind in the environme nt and performance of the simulator. The natural frequency of each system was determined by applying a 20 lb force to the signal and then recording the rotations after releasing. The primary mode for cable supported signals is in the vertical direction. T he primary frequencies for all systems were around 0.5 cycles per second. Frequencies in other directions were two to five times higher than the primary frequencies. Only the primary frequency was used during testing.

PAGE 37

37 Figure 3 1 Spacing and layout of cables

PAGE 38

38 Figure 3 2 Cable support systems: A) Dual Cable Support System; B) Discontinuous Messenger Cable System; C) Single Cable System

PAGE 39

39 Figure 3 3 The UF Hu (Photo courtesy of author) Figure 3 4 (Photo courtesy of author)

PAGE 40

40 Figure 3 5 Signal and simulator testing orientations Figure 3 6 Two signal test setup (Photo courtesy of author)

PAGE 41

41 Figure 3 7 R.M. Young anemometer (Photo courtesy of author) Figure 3 8 Load cells installed in line with catenary and messenger cables (Photo courtesy of author)

PAGE 42

42 Figure 3 9 String potentiometer setup Figure 3 10 Mounting system for string pot entiometers (Photos courtesy of author)

PAGE 43

43 Figure 3 11 Orientation sensor installed inside the signal (Photo courtesy of author)

PAGE 44

44 Figure 3 12 Types of hangers: A) Pipe Hanger; B) Adjustable H anger; C) Cable Hanger; D) Pivotal Hanger; E) Direct Connection (Photo s courtesy of author) Figure 3 13 Sample of oscillating wind load sequence (Test 4)

PAGE 45

45 Figure 3 14 Sample of linear wind load (Test 1)

PAGE 46

46 Table 3 1 Test Schedule Test no. Wind angle () Signal angle () Cable system Hanger/ connection % sag No. of signals Date Series no. 1 90 90 Dual Pipe 5 1 1/20/2011 1 2 90 90 Dual Pivotal 5 1 1/20/2011 1 3 90 90 Single Direct connect. 3 1 1/20/2011 1 4 90 90 Dual Pipe 5 1 4/28/2011 2 5 90 90 Dual Pivotal 5 1 4/28/2011 2 6 90 90 Dual Cable 5 1 4/28/2011 2 7 90 90 Single Direct connect. 3 1 4/28/2011 2 8 90 90 Single Direct connect. 3 1(poly.) 4/28/2011 2 9 45 90 Dual Pipe 5 1 5/1/2011 3 10 45 45 Dual Pipe 5 1 5/1/2011 3 11 45 90 Dual Pivotal 5 1 5/1/2011 3 12 45 45 Dual Pivotal 5 1 5/1/2011 3 13 45 90 Dual Cable 5 1 5/1/2011 3 14 45 45 Dual Cable 5 1 5/1/2011 3 15 45 90 Single Direct connect. 3 1 5/1 /2011 3 16 45 45 Single Direct connect. 3 1 5/1/2011 3 17 45 90 Dual Pipe 5 2 5/10/2011 4 18 45 90 Dual Pivotal 5 2 5/10/2011 4 19 45 90 Dual Cable 5 2 5/10/2011 4 20 45 90 Disc. mess. Adj. strap 5 2 5/10/2011 4 21 45 90 Single Direct connect. 3 2 5 /10/2011 4 22 45 90 Single Direct connect. 3 1(poly.) 5/11/2011 5 23 45 90 Dual Adj. strap 5 1 5/11/2011 5 24 45 90 Dual Pivotal** 5 1 5/11/2011 5 25 45 90 Dual Pivotal 5 1(poly.) 5/11/2011 5 26 45 90 Dual Cable 5 1(poly.) 5/11/2011 5 27 45 90 Dual Pipe ** 5 1 5/11/2011 5 28 12 90 Dual Pipe 5 1 5/18/2011 6 29 12 90 Dual Pivotal 5 1 5/18/2011 6 30 12 90 Dual Cable 5 1 5/18/2011 6 31 12 90 Dual Pivotal 5 1(poly.) 5/18/2011 6 32 12 90 Single Direct connect. 3 1(poly.) 5/18/2011 6 33 12 90 Single Direct connect. 3 1 5/18/2011 6 *Linearly ramped wind load and 40" cable separation **Messenger cable clamped to front side of hanger instead of back side.

PAGE 47

47 CHAPTER 4 TEST RESULTS Tests for this project were performed from January 20, 2011, to May 18, 2011. Representatives from the Florida Department of Transportation (FDOT), manufacturers of the equipment being tested, and students and advisors from the University of Florida (UF) were present to view the testing. The key interests for the collection of data were the rotations experienced by the traffic signals, the tension in the catenary and messenger cables, and the displacement of the cables. 4.1 Signal Rotations The wide coastal exposure and frequency of tropical storms and hurricanes that occur i n Florida make the state susceptible to frequent high winds. In addition to the concern of damage to traffic signals in high winds, serviceability related to loss of signal visibility from signal rotation is a concern at lower level winds. When a signal is blown backwards, it loses visibility at a vertical (backwards) rotation of 27.5 (Institute of Transportation Engineers (ITE) 2005) ( Figure 2 5 ). In all tests, the horizontal (sideways) rotation limit was not met or was exceeded after passing the limit for vertical rotation. The ITE visibility limit shown in the figures in this section is 27.5 because the backward vertical rotation of the signal controls the limit. 4.1.1 Effect of Signal and Simulator Orientations on Signal Rotation The single cable support system was tested at all of the orientations shown in Figure 3 5 to serve as a baseline to compare other systems. The pipe, pivotal, and cable hangers were also tested a t all orientations. As seen in Figure 4 1 when the wind load was below 45 mph, the single cable system experienced the most rotation when at a 90 signal orientation and a 90

PAGE 48

48 simulator orientation ( Figure 3 5 A). This provided the maximum exposed area to wind, and did not restrict the backward rotation of the signal. The single cable system experienced continuous swinging motions above 65 mph wind loads for all orientation s. Figure 4 2 to Figure 4 4 show that the 90 signal and 90 simulator orientation caused the most rotation for the dual cable systems. The pivotal and cable hangers were n ot as sensitive to orientation as the pipe hangers when wind speeds exceeded 60 mph. Because the dual cable systems experienced less erratic rotations than the single cable system, the figures show cleaner differences in the effects of orientation on rotat ion. The pivotal hanger only deviated from typical rotation patterns at the 90 signal and 45 simulator orientations. 4.1.2 Effect of Hanger Types and Support Systems on Signal Rotation The hangers used with the dual cable support system varied in rigidi ty and flexibility. The standard drop pipe hanger is constructed from a single rigid pipe. The adjustable strap hanger is constructed from two rigid pieces bolted together at two points and may be extended with the addition of a flat bar bolted between the two base pieces. The pivotal hanger consists of two rigid portions attached by a hinge which allows free forward and backward rotation when ignoring the restriction of the messenger cable. The pivotal hanger may also be made larger with the addition of a flat bar bolted between the top and bottom pieces. The cable hanger is flexible along the wire strand cable was used for the cable hanger; however, only the diameter was specified and not the type of cable. Other cables may hav e more or less flexibility which could change the behavior of the system. Test Series 2 included the single cable support system with a direct connection and the dual cable support system with the pipe, pivotal, and cable hangers tested at

PAGE 49

49 90 signal and 90 simulator orientations. Figure 4 5 shows that pivotal and cable hangers with the dual cable system experienced nearly identical rotations. The pipe hanger prevents the distance between the messenger and catenar y cables from changing because of the lack of a hinge or flexibility. The high messenger tension also created a resistance to rotation. The single cable system performed the same as the pivotal and cable hangers until wind speeds exceeded 65 mph. Then the single cable system behaved more like the pipe hanger. At the 90 signal and 90 simulator orientation, the pipe hanger, cable hanger, and single cable system exceeded the ITE visibility limit with a wind velocity of approximately 36 mph. The pipe hanger v isibility limit occurred above a 40 mph wind velocity. The adjustable strap hanger was added to the test schedule at the 90 signal and 45 simulator orientations. The adjustable strap hanger performed much like the pipe hanger, but it yielded during testi ng ( Figure 4 7 ), which may have caused the higher rotations. At the maximum wind speed, there was approximately a 10 difference in rotations between the strap and pipe hangers. At less than 60 mph wind speeds, the difference was negligible between these two hangers. The pivotal and cable hangers performed again with little difference at this orientation. For the pivotal and cable hangers, the ITE visibility limit was met around 45 mph winds, and for the pipe and st rap hangers it was met around 50 mph. For all systems, the ITE visibility limit occurred during the oscillating wind loads, so that visibility was lost within a range of wind velocities. At the 45 signal and 45 simulator orientation, the pivotal hanger n o longer behaved like the cable hanger. Because exposed signal area and wind force were

PAGE 50

50 maximized in a direction not parallel with the direction of the hinge on the pivotal hanger, the rotation was significantly reduced. Fig ure 4 8 shows that for wind speeds below 60 mph, the pivotal hanger behaved like a rigid pipe hanger. Above 60 mph, the rotation was further restricted, and the signal rotation with the pivotal hanger was less than with the pipe hanger. The hinge on the pivotal hanger only frees motion perpendicular to the cable span. At this angle, the signal blew against the hinge. During the final test series, there was a 90 signal orientation and 12 simulator orientation. This combination had the smallest si gnal area exposure to the wind. The pipe hanger experienced the lowest rotations ( Figure 4 9 ). The other systems performed with similar rotations to each other throughout the loading sequence. The results at this o rientation showed more swinging of the signals during loading, especially at wind speeds higher than 60 mph. The discontinuous messenger cable system is an alternative cable support system that requires two or more signals. Test Series 4 repeated tests on the dual cable system with the pipe, pivotal, and cable hangers, and with the single cable system at the 90 signal and 45 simulator orientation using two signals in order to compare them with the discontinuous messenger cable system. Figure 4 10 and Figure 4 11 show the rotation results for the two signal tests. The rotations were measured in both the 5 section signal and the 3 section signal. The severed messenger cable used in the discontinuous messenger cable system prevents the unpredictable rotations that are found in the single cable system. When the wind speeds were less than 60 mph, the discontinuous messenger cable system had rotation behavior similar to the dual cable system with pivotal and cable hangers and the single cable system Above 60 mph wind speeds, the

PAGE 51

51 rotation in the discontinuous messenger system leveled off until it matched the rotations of the pipe hanger system. The effects of having multiple signa ls on a single span are discussed in the next section. Figure 4 10 shows that for the 5 section signal the discontinuous messenger cable system and the cable hanger had a visibility limit at 40 mph, the single cabl e system and the pivotal hanger had a visibility limit of 44 mph, and the pipe hanger had a visibility limit between 50 mph and 55 mph. The 3 section signal shows less difference among the systems. 4.1.3 Effect of Multiple Signals on Signal Rotation At th e 45 simulator and 90 signal orientation, the dual cable system with pipe, pivotal, and cable hangers, and the single cable system were retested with a 3 section signal added to the span. The two signals were adjusted along the span to remain completely within in the wind field. There was a five foot center to center separation of the signals. The single cable system experienced the most significant effects of adding a second signal ( Figure 4 12 ). Signal rotation s were more irregular as the wind load increased when an additional signal was added to the span. The 5 section signal during the two signal test experienced more rotation compared to the one signal test, while the 3 section signal experienced less rotatio n compared to the one signal test. At less than 45 mph wind speeds, there appeared to be negligible difference in the rotation results. The pipe hanger and pivotal hanger showed no change in rotation when an additional signal was added to the span ( Figure 4 13 and Figure 4 14 ). When a second signal was added to the cable hanger system, the 5 section signal experienced higher rotations ( Figure 4 15 ). The 3 section signal experienced rotations similar to the one signal test.

PAGE 52

52 4.1.4 Effect of Signal Weight on Signal Rotation Various tests were repeated using a polycarbonate signal in place of the aluminum signal. The aluminum signal weig hed 73.0 lbs, and the polycarbonate signal weighed 53.5 lbs. As well as the difference in weights, the aluminum signal had a front area of 14.2 square feet, while the polycarbonate signal had a front area of 13.1 square feet. The polycarbonate signal consi sted of an aluminum top signal head and four polycarbonate signal heads. All five signal heads are aluminum for the aluminum signal. The single cable support system was tested with the polycarbonate signal at the 90 signal and 90 simulator orientation, t he 90 signal and 45 simulator orientation, and the 90 signal and 12 simulator orientation. The results in Figure 4 16 and Figure 4 17 show erratic behavior for the poly carbonate signal at the 90 signal and 90 simulator orientation and at the 90 signal and 45 simulator orientation. The backplate of the polycarbonate signal broke off before the wind speed reached 110 mph ( Figure 4 16 ). The reduced area exposed to the wind dampened the swinging of the signal. With the exposed area of the signal reduced at the 90 signal and 12 simulator orientation, the polycarbonate signal experienced stable rotation ( Figure 4 18 ). When using the polycarbonate signal with the dual cable system, all of the results maintained stable behavior ( Figure 4 19 to Figure 4 21 ). Th ese tests were conducted using the pivotal hanger and the cable hanger at the 90 signal and 45 simulator orientation and with the pivotal hanger at the 90 signal and 12 simulator orientation. For all of these tests, the polycarbonate signal experienced higher rotations when wind speeds exceeded 45 mph.

PAGE 53

53 4.1.5 Effect of Messenger Cable Clamp Location on Signal Rotation Tests 9 and 11 were repeated with the same orientation, but installing the messenger clamps on the front of the pipe and pivotal hangers. Figure 4 22 and Figure 4 23 show that there is very little difference in each case. After exceeding 60 mph, the rotations were only slightly higher for the case where the messenger was clamped to the back. The fall in rotation at the end of Test 9 occurred after a repeatedly used backplate failed at one corner. 4.2 Cable Tensions The tension experienced in the supporting cables is important for calculating moment and str esses experienced in the strain poles. During installation for testing, the workers from the Gainesville Traffic Operations were asked to replicate field practices. Each signal was first hung on the catenary cable, and then the messenger cable was tightene d across the span before attaching it to the hanger. Except for the first test series, the tension of the messenger cable was controlled so that all initial messenger tensions were between 750 lbs and 1200 lbs. The sag for all dual cable systems was 5% and for single cable systems was 3%. Table 4 1 through Table 4 5 summarize the changes in tension during each test from Test Series 2 through 6. The initial tension for each cable and the change in tension at wind speeds of 60 mph and 115 mph are provided. These speeds were selected because in the midrange speeds, behavior of each system was more distinct, and because some tests ended before the wind speeds reached 120 mph. No te that the tensions reported in these sections are the change in tension during testing. This created a common starting point for each data set which was necessary because initials tensions are not equal on all tests.

PAGE 54

54 For all tests, the catenary cables l ost tension as wind speeds increased until approximately 60 mph. At that point, some systems started to recover tension. All of the changes in catenary tension remained very small. The messenger cables for all dual cable systems increased in tension as win d speeds increased. The final change in tension was very high in messenger cables. 4.2.1 Effect of Signal and Simulator Orientations on Cable Tension Changing the orientation of the signal and simulators had various effects on the cable tensions. Figure 4 24 compares the results for all of the single cable system tests that had no special conditions. Similar to the results for rotation, the 90 signal and 90 simulator orientation caused the highest final tensions for the single cable system. The 90 signal and 12 simulator orientation was the only case for the single cable system that did not regain the full initial tension during testing. The catenary cable for the dual cable system with a pipe hanger experience d similar results as the single cable support. The 90 signal and 90 simulator orientation caused the highest change in catenary tension, and the 90 signal and 12 simulator orientation caused the lowest change in catenary tension ( Figure 4 25 ). The messenger cable shows more exaggerated effects of changing the orientations of the signal and hurricane simulator. The tensions of the messenger cables all increased throughout testing. The difference became as high as 1500 lbs greater than the initial tension for the 90 signal and 90 simulator orientation and 90 signal and 45 simulator orientation at 120 mph. At the 90 signal and 12 simulator orientation, the limited area of exposure prevented excessive gains in tension and the final change in messenger tension was approximately 700 lbs.

PAGE 55

55 The tensions for the pivotal hanger and the cable hanger were less affected by orientation ( Figure 4 26 and Figu re 4 27 ). The catenary cables for these systems did not regain the lost tension as wind speeds increased because these cables became inactive with the system. The final differences for messenger cable tension ranged from 750 lbs to 1000 lbs for the pivot al hanger and from 500 lbs to 1000 lbs for the cable hanger. 4.2.2 Effect of Hanger Types and Support Systems on Cable Tension The complete rigidity of the pipe hanger caused it to maintain the separation distance between the catenary and messenger cables As the signal rotated, the cables were forced outward causing the support cables with the pipe hanger to maintain the highest changes in tension in most cases. The freedom of rotation allowed by the pivotal and cable hangers significantly reduced the ten sion increase experienced in the dual cable system. For the 90 signal and 90 simulator orientation, the messenger cable for the pipe hanger had a significantly higher change in tension while for the pivotal and cable hangers had the same changes in tens ion throughout testing at this orientation ( Figure 4 28 ) The same conclusions were found for the 90 signal and 45 simulator orientation ( Figure 4 29 ). The adjustable strap hanger was also tested at thi s orientation. The strap hanger performed most like the pipe hanger, but the tensions were slightly less. This case yielded the strap hanger, which may have caused the tension to experience less change. At the 45 signal and 45 simulator orientation, the single cable system experienced sharp fluctuations in the catenary tension at high wind speeds ( Figure 4

PAGE 56

56 30 ). The cable hanger had the lowest change in catenary and messenger tension. The pipe hanger had the highe st change in messenger tension throughout testing. The 90 signal and 12 simulator orientation did not provide adequate differences to make comparisons among the various systems ( Figure 4 31 ). All results were wi thin a 250 lb range, and most results overlapped. Where results did not overlap, the previous patterns were repeated. Test Series 4 consisted of tests with two signals. These tests showed that the use of a discontinuous messenger cable provided more stabl e tension results than the single cable system without causing additional forces on strain poles by the attachment of a messenger cable ( Figure 4 32 ). The catenary tension of the discontinuous messenger cable syste m decreased more than for the single cable system or the dual cable system with a pipe hanger. The remaining systems continued to perform with the same patterns as with one signal. 4.2.3 Effect of Multiple Signals on Cable Tension All tests with two signa ls were performed at the 90 signal and 45 simulator orientation. The changes in tension for all systems were greater when a second signal was added to any system. When a second signal was added to the cable span of the single cable system, the tension ch anges developed unpredictable fluctuation at winds speeds higher than 60 mph ( Figure 4 33 ). The mean change in tension followed the same shape as with one signal, but the fluctuation caused tensions to reach as high as 1000 lbs ad ditional force, which is comparable to the forces developed by messenger cables in dual cable systems.

PAGE 57

57 For all of the dual cable systems, the catenary and messenger tensions underwent greater changes with a second signal ( Figure 4 34 to Figure 4 36 ). The pivotal and cable hangers were more affected by the second signal with changes up to 500 lbs higher than with one signal. The change in messenger tension with the pipe hang er was up to 250 lbs higher. 4.2.4 Effect of Signal Material on Cable Tension The polycarbonate signal and the aluminum signal had a 20 lb weight difference. The use of the polycarbonate signal with the single cable system caused the catenary tension to be come very unpredictable in most orientations ( Figure 4 37 and Figure 4 38 ). With the polycarbonate signal at the 90 signal and 90 simulator orientation, the change in ten sion fluctuated between 250 lbs and 2250 lbs before the backplate failed at 107 mph wind speeds. After the failure of the backplate, change in tension matched that of the aluminum signal. With the 90 signal and 45 simulator orientation, the polycarbonat e signal experienced tension fluctuations between 250 lbs and 1250 lbs. At the 90 signal and 12 simulator orientation, the backplate provided much less exposed area and effect of the signal weight was insignificant ( Figure 4 39 ). The pivotal and cable hangers did not experience significant effects in cable tension with weight change. The messenger cable prevented the irregular fluctuation in tension. 4.2.5 Effect of Messenger Clamp Location on Cable Tension The location of the messenger clamp for the pipe and pivotal hangers had little effect on the cable tensions ( Figure 4 40 and Figure 4 41 ). The pipe hanger lost less catenary tension when the messenger cable was clamped to the back side of the hanger than when clamped to the front, but the messenger cable experienced no

PAGE 58

58 difference with change in clamp location. The pivotal hanger also experienced smaller changes in catenary and messenger tension with a back clamp instead of a front clamp. 4.3 Cable Displacements The displacements in the catenary and messenger cables were recorded during testing. Some string pots malfunction ed during some tests. Periodically, the data was not re ceived from the wireless transmitter for the string pots, and during one test a cord was cut by a failed backplate. The following figures show data from Test Series 2. These tests provided complete data sets. The remaining unaffected data can be found in A ppendix C. F or dual cable systems the messenger cables experienced displacements within a range that was consistent with the elastic behavior of the cables for the cable tension measured. Across the 50 foot length of span wire there was less than a quarter inch difference between the cable elongation based off measured displacements and the elastic elongation calculated from the change in tension. The slight differences were assumed to be caused by slipping of the cables at the clamps on the strain poles. When testing the pipe hanger, the signal rotation caused the messenger cable to displace backwards. Because of the rigidity of the pipe, the catenary cable was forced to displace forward. The forward displacement of the catenary cable had a five inch maxim um. The maximum displacement of the messenger cable was 13 inches. The pivotal and cable hangers allowed for independent movement of the catenary cable from the messenger cable after the signal weight was removed from the catenary cable by wind lift and tr ansferred to the messenger cable. For both of these systems, the signal pulled the messenger cable into backwards displacement. The displacement caused by the wind on the catenary cable was much smaller ( Figure 4 43 and Figure 4

PAGE 59

59 44 ). For the pivotal hanger, the maximum catenary cable displacement was six inches and the maximum messenger cable displacement was 12 inches. For the cable hangers, the maximum catenary cable displ acement was 8 inches and the maximum messenger cable displacement was 12 inches. The single cable system experienced the greatest cable displacement. Without the restriction given by the messenger cable on dual cable systems, the single cable swung back fr eely under wind loading ( Figure 4 44 ). The maximum displacement was approximately 26 inches.

PAGE 60

60 Figure 4 1 Rotation for single cable at all orientations (Tests 7, 15, 16, and 33)

PAGE 61

61 Figure 4 2 Rotation for pipe hanger at all orientations (Tests 4, 9, 10, and 28) Figure 4 3 Rotation for pivotal hanger at all orientations (Tests 5, 11, 12, and 29)

PAGE 62

62 Figure 4 4 Rotation for cable hanger at all orientations (Tests 6, 13, 14, and 30)

PAGE 63

63 Figure 4 5 Rotation for systems at 90 signal and 90 simulator orientation (Test Series 2)

PAGE 64

64 Figure 4 6 Rotation for systems at 90 signal and 45 simulator orientations (Tests 9, 11, 13, 15, and 23) Figure 4 7 Adjustable strap hanger with visible yielding in lower section (Test 23) (Photo courtesy of author)

PAGE 65

65 Fig ure 4 8 Rotation for systems at 45 signal and 45 simulator orientations (Tests 10, 12, 14, and 16)

PAGE 66

66 Figure 4 9 Rotation for systems at 90 signal and 12 simulator orientations (Tests 28, 29, 30, and 33)

PAGE 67

67 Figure 4 10 Rotation for 5 section signals during Test Series 4 two at 90 signal and 45 simulator orientations

PAGE 68

68 Figure 4 11 Rotation for 3 section signals during Test Series 4 two at 90 signal and 45 simulator orientations

PAGE 69

69 Figure 4 12 Rotation for single cable system with two signals verses one signal (Tests 15 and 21)

PAGE 70

70 Figure 4 13 Rotation for pipe hanger wi th two signals verses one signal (Tests 9 and 17)

PAGE 71

71 Figure 4 14 Rotation for pivotal hanger with two signals verses one signal (Tests 11 and 18)

PAGE 72

72 Figure 4 15 Rotation for cable hanger with tw o signals verses one signal (Tests 13 and 19)

PAGE 73

73 Figure 4 16 Rotation for single cable system with aluminum signal verses polycarbonate signal at 90 signal and 90 simulator orientation (Tests 7 and 8)

PAGE 74

74 Figure 4 17 Rotation for single cable system with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 15 and 22)

PAGE 75

75 Figure 4 18 Rotation for single cable system with aluminum ve rses polycarbonate signal at 90 signal and 12 simulator orientation (Tests 33 and 32)

PAGE 76

76 Figure 4 19 Rotation for pivotal hanger with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 1 1 and 25)

PAGE 77

77 Figure 4 20 Rotation for pivotal hanger with aluminum verses polycarbonate signal at 90 signal and 12 simulator orientation (Tests 29 and 31)

PAGE 78

78 Figure 4 21 Rotation for cable han ger with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientation (Tests 13 and 26)

PAGE 79

79 Figure 4 22 Rotation for pipe hanger with altered messenger clamp location (Tests 9 and 27)

PAGE 80

80 Figure 4 23 Rotation for pivotal hanger with altered messenger clamp location (Tests 11 and 24)

PAGE 81

81 Figure 4 24 Change in tension for single cable at all orientations (Tests 7, 15, 16, and 33)

PAGE 82

82 Figure 4 25 Change in tension for pipe hanger at all orientations (Tests 4, 9, 10, and 28) Messenger Cables Catenary Cables

PAGE 83

83 Figure 4 26 Change in tension for pivotal hanger at all orientations (Tests 5, 11, 12, and 29) Messenge r Cables Catenary Cables

PAGE 84

84 Figu re 4 27 Change in tension for cable hanger at all orientations (Tests 6, 13, 14, and 30) Messenger Cables Catenary Cables

PAGE 85

85 Figure 4 28 Change in tension for all systems at 90 signal and 90 simulator orientations Messenger Cables Catenary Cables

PAGE 86

86 Figure 4 29 Change in tension for all systems at 90 signal and 45 simulator orientations Messenger Cables Catenary Cables

PAGE 87

87 Figure 4 30 Change in tension for all systems at 45 signal and 45 simulator orientations Messenger Cables Catenary Cables

PAGE 88

88 Figure 4 31 Change in tension for all systems at 90 signal and 12 simulator orientations Messenger Cables Catenary Cables

PAGE 89

89 Figure 4 32 Change in tension for all systems with two signals Messenger Cables Catenary Cables

PAGE 90

90 Figure 4 33 Change in tension for single cable system with two verses one signal (Tests 15 and 21)

PAGE 91

91 Figure 4 34 Change in tension for pipe hanger with two verses one signal (Tests 9 and 17) Messenger Cables C atenary Cables

PAGE 92

92 Figure 4 35 Change in tension for pivotal hanger with two verses one signal (Tests 11 and 18) Messenger Cables Catenary Cables

PAGE 93

93 Figure 4 36 Change in tension for cable hanger with two signals verses one signal (Tests 13 and 19) Messenger Cables Catenary Cables

PAGE 94

94 Figure 4 37 Change in tension for the single cable system with aluminum verses polycarbonate signal at 90 signal and 90 simulator orientations (Tests 7 and 8)

PAGE 95

95 Figure 4 38 Change in tension for the s ingle cable system with aluminum verses polycarbonate signal at 90 signal and 45 simulator orientations (Tests 15 and 22)

PAGE 96

96 Figure 4 39 Change in tension for the single cable system with aluminum verses polycarbonate signa l at 90 signal and 12 simulator orientations (Tests 33 and 32)

PAGE 97

97 Figure 4 40 Change in tension for pipe hanger with altered messenger clamp location (Tests 9 and 27) Messenger Cables Catenary Cables

PAGE 98

98 Figure 4 41 Chan ge in tension for pivotal hanger with altered messenger clamp location (Tests 11 and 24) Messenger Cables Catenary Cables

PAGE 99

99 Figure 4 42 Cable displacement for pipe hanger at 90 signal and 90 simulator orientation

PAGE 100

100 Figure 4 43 Cable displacement for pivotal hanger at 90 signal and 90 simulator orientation

PAGE 101

101 Figure 4 44 Cable displacement for cable hanger at 90 signal and 90 simulator orientation

PAGE 102

102 Figure 4 45 Cable displacement for single cable system at 90 signal and 90 simulator orientation

PAGE 103

10 3 Table 4 1 Changes in cable tension for Test Series 2 Test no. Hanger Wind/ signal angles () Cable Initial tension (lbs) Tension at 60 mph (lbs) Tension change at 60 mph (lbs) Tension at 115 mph (lbs) Tension change at 115 mph (lbs) Notes 4 Pipe 90/90 Cat. 429 416 13 535 106 Mess. 835 1622 787 2324 1489 5 Pivotal 90/90 Cat. 415 280 135 115 300 Mess. 1179 1565 38 6 2094 915 6 Cable 90/90 Cat. 394 179 215 130 264 Mess. 1167 1535 368 2076 909 7 Direct connect 90/90 Cat. 572 461 111 577 5 8 Direct connect 90/90 Cat. 468 356 112 553 85 Poly. signal Table 4 2 Change s in cable tension for Test Series 3 Test no. Hanger Wind/ signal angles () Cable Initial tension (lbs) Tension at 60 mph (lbs) Tension change at 60 mph (lbs) Tension at 115 mph (lbs) Tension change at 115 mph (lbs) Notes 9 Pipe 45/90 Cat. 398 334 64 476 78 Mess. 817 1436 619 2318 1501 10 Pipe 45/45 Cat. 266 208 58 314 48 Mess. 1174 1636 462 2515 1341 11 Pivotal 45/90 Cat. 297 142 155 151 146 Mess. 875 1317 442 1645 770 12 Pivotal 45/45 Cat. 299 135 164 221 78 Mess. 891 1312 421 1588 697 13 Cable 45/45 Cat. 279 114 165 46 233 Mess. 889 1230 341 1599 710 14 Cable 45/45 Cat. 277 99 178 58 219 Mess. 858 1163 305 1451 593 15 Direct connect 45/90 Cat. 540 475 65 695 155 16 Direct connect 45/45 Cat. 547 395 152 765 218

PAGE 104

104 Table 4 3 Changes in cable tension for Test Series 4 Test no. Hanger Wind/ signal angles () Cable Initial tension (lbs) Tension at 60 mph (lbs) Tension change at 60 mph (lbs) Te nsion at 115 mph (lbs) Tension change at 115 mph (lbs) Notes 17 Pipe 45/90 Cat. 584 468 116 584 0 2 signals; Mess. 1102 1949 847 2780 1678 18 Pivotal 45/90 Cat. 567 352 215 286 281 2 signals; Mess. 866 1514 648 2043 1177 19 Cable 45/90 Cat. 557 299 258 141 416 2 signals; Mess. 767 1409 642 1875 1108 20 A dj. strap 45/90 Cat. 571 457 114 459 112 2 signals; disc. mess. 21 Direct connect 45/90 Cat. 863 775 88 897 34 2 signals; Table 4 4 Changes in cable tension for Test Series 5 Test no. Hanger Wind/ signal angles () Cable Initial tension (lbs) Tension at 60 mph (lbs) Tension change at 60 mph (lbs) Tension at 115 mph (lbs) Tension change at 115 mph (lbs) Notes 22 Direct connect 45/90 Cat. 433 406 27 654 221 Poly. signal 23 A dj. strap 45/90 Cat. 409 321 88 376 33 Mess. clamped to front Mess. 1083 1568 485 2296 1213 24 Pivotal 45/90 Cat. 419 260 159 204 215 Mess. clamped to front Mess. 871 1339 468 1652 781 25 Pivotal 45/90 Cat. 350 214 136 190 160 Poly. signal Mess. 896 1222 326 1571 675 26 Cable 45/90 Cat. 328 211 117 130 198 Poly. signal Mess. 881 1151 270 1455 574 27 Pipe 45/90 Cat. 418 312 106 457 39 Mess. clamped to front Mess. 841 1410 569 2277 1436

PAGE 105

105 Table 4 5 Changes in cable tension for Test Series 6 Test no. Hanger Wind/ signal angles () Cable Initial tension (lbs) Tension at 60 mph (lbs) Tension change at 60 mph (lbs) Tension at 115 mph (lbs) Tension change at 115 mph (lbs) Notes 28 Pipe 12/90 Cat. 324 274 50 220 104 Mess. 1148 1504 356 1789 641 29 Pivotal 12/90 Cat. 329 193 136 222 107 Mess. 1000 1315 315 1674 674 30 Cable 12/90 Cat. 327 214 113 77 250 Mess. 955 1187 232 14 61 506 31 Pivotal 12/90 Cat. 251 159 92 186 65 Poly. signal Mess. 1015 1291 276 1618 603 32 Direct connect. 12/90 Cat. 441 364 77 421 20 Poly. signal 33 Direct connect. 12/90 Cat. 563 471 92 431 132

PAGE 106

106 CHAPTER 5 FORCE COEFFICIENTS F OR WIND FORCES The data collected during this project was used to calculate force coefficients for the calculation of wind forces on traffic signals. Drag and lift coefficients for single cable systems were calculated by Cook and J ohnson These were appropriate for the single cable system since both the cable and signal rotated in the wind. For the dual ca ble system, the mess enger cable provided a horizontal restraint that pre vented s way of the system. As a result, for the dual cable system, a lateral force coefficient for evaluating the reaction of the messenger cable was appro priate The following information was determined using the 90 signal and 90 simulator orientation. The resultant force on the messenger cable caused by the wind force was assumed to be a lateral force for the dual cable system, unlike for the single cable system where the resultant force is depe ndent on the rotation of the plane of the cable ( Figure 5 1 ). In order to calculate the resultant force on the messenger cable, a modification of Equation 2 4 was used (Eq. 5 1). (5 1) The deflection of the mes senger cable ( D M ) and the change in messenger cable tension ( T M ) were used in place of the sag and actual tension from Eq. 2 4, which was for the single cable system Another factor left off of Eq. 2 4 was the weight of the cable because only the change in tension was considered. The force coefficient was determined with equations similar to those for the drag and lift coefficients found by Cook and Johnson. The force coefficient is a ratio of the resultant force on the signal to the wind force on the are a of the signal (Eq. 5 2).

PAGE 107

107 (5 2) The catenary cable forces were ignored when considering the dual cable systems because for all cases, the change in tension in the catenary cables was not significant. It was assumed that the catenary cable did not aff ect the resultant force on the traffic signal. The force coefficient may be applied to Equation 5 3, where P w is the wind force in pounds, V is the wind velocity in miles per hour, and A is the area of the object undergoing wind loading in square feet ( Figure 5 2 ) (5 3) Figure 5 3 and Figure 5 4 show the force coefficients with respect to signal rotation and to wind velocity, respectively. The maximum force coefficient for the pipe hanger was 0.4 5 while for the pivotal and cable hangers it was 0.2 0 The area was not varied with rotation for these calculations. For design purposes, the maximum force coefficients are recommended for use. The pipe hanger, which restricted rotation of the signal, had a much higher force coefficient than the pivotal and cable hangers, which allowed rotation of the signal to occur. This reinforces what was seen from the signal rotations and changes in tension experien ced by these systems. These force coefficients are lower than the current recommended drag coefficients from the American Association of State Highway Transportation Officials (AASHTO) and the Florida Department of Transportation (FDOT). As found by Alampa lli in 1997, the recommended values create conservative loads for strain poles.

PAGE 108

108 Figure 5 1 Free body diagrams for traffic signals with A) single cable system, B) pipe hanger, and C) pivotal and cable hangers Figure 5 2 Top view of wind force applied to signal with reactions in messenger cable

PAGE 109

109 Figure 5 3 Force coefficient vs. signal rotation for dual cable systems Figure 5 4 Force coefficient vs. wind velocity for dual cable systems

PAGE 110

110 CHAPTER 6 PERFORMANCE OF EQUIP MENT AND HARDWARE While testing the various support systems, most of the equipment was reused in several tests. A few problems that were encountered included rapid wearing of bolts, fatigue in signal visors and backplates at connections, and other failures that were not part of the research focus. Appendix D contains a catalog of equipment that includes photographs and lists of tests each piece of equipment was used for. The first failure was encountered during Test Series 1. Schedule 40 aluminum was initially used for the pipe hangers, but the material failed before reaching the target wind speed of 120 mph. The failure occurred at the point where the pipe was threa ded for an adapter to fit the disconnect box. The test was repeated with the same material in order to verify that the pipe was properly threaded and installed. When the pipe failed a second time, it was decided to make some changes to the testing schedule in order to maximize the usable data collected without failing the equipment. Schedule 80 aluminum was used for the remaining tests with pipe hanger. Test 8 was the first to use a polycarbonate signal. During this test, it was discovered that the polycar bonate signal developed very erratic behavior at high winds speeds. The signal was tossed very violently by the wind during this test. The backplate was torn off at approximately 107 mph. The top head of the polycarbonate signal also became loose and twist ed slightly in the bracket joining it to the other heads ( Figure 6 1 ) During Test 22, the backplate was again torn away by the wind. A 72 teeth serrated boss was used to join the top head and the two way bracket The serrated bos s typically prevents twisting at the connection During Test 22 t he teeth were completely worn

PAGE 111

111 down by the parts twisting from the wind force ( Figure 6 2 ). The connection for the backplate also broke away from the signal ( Figure 6 3 ). During Test 8, the two way bracket for the polycarbonate signal was attached with 1/4 inch bolts and 7/16 inch nuts. Upon examination it was noticed that the smaller tri stud nuts had loosened which allowed the top section to become detached from the two way bracket serrated teeth allowing twisting During Test 22, the 1/4 inch bolts and 7/16 inch nuts were replaced with 5/16 inch bolts and 1/2 inch nylon insert lock nuts. The same type of failure occurred for this test, bu t the nuts appeared to have not loosened. However, the tri stud washer had yielded and bec a me concave which loosened the connection and allowed the signal head to rotate and damage both the serrated edge of the signal and two way bracket. The FDOT determin ed that the tri stud bolts should have a 5/16 inch minimum diameter and that the thickness or strength of the tri stud washer needs to be increased. An additional modification would be to increase the depth of the serrated teeth on the signal and two way b racket. It is also recommended that the mounting hardware manufacturers provide all connection hardware with their brackets. While the backplate failure in Test 8 was due to the behavior of the equipment being used, the backplate failure in Test 27 was lik ely due to overuse of the same backplate. As seen in Figure 6 4 the backplate material yielded during Test 27 after several uses. The failure in Test 8 was similar to this. Several bolts had to be cut away while changing equipmen t. After a couple of uses, the messenger cable clamp on a pipe hanger could not be reused because wear in the threads caused the nuts to become stuck to the U bolt. A similar issue occurred

PAGE 112

112 with the bolt at the base of the pivotal hangers. Many washers, in cluding tri stud washers, had to be replaced as well due to over tightening during assembly and installation. The visors were replaced several times during testing. The small hooks for attaching the visors would typically fail on either side at the bott om of the signal head. There was a visor on all five heads of the 5 section signals and all three heads of the 3 section signals. The visor failures occurred once or twice for every series of tests, but the failures were minor and most likely had no effect on test data because only one or two of the four hooks would break and the visor would stay attached. The adjustable strap hanger that was used in Test 23 experienced yielding. This was expected because in previous research, strap hangers consistently yi elded at high wind speeds. This behavior was not seen while using the discontinuous messenger cable system. The original stabilizer clamp used during testing cracked in the center of the clamp after multiple tests. This was a sand cast aluminum part that w as approximately two inches wide and six inches long ( Figure 6 5 ). After breaking the casted piece, an extruded aluminum part was used as a replacement. The extruded aluminum appeared to be more durable. This extruded stabilizer c lamp did not experience any crack or breakages during the remaining tests. The FDOT has requested that manufacturers improve the part by using of extruded aluminum instead of sand cast aluminum for this product, shortening the length of the clamp, or compl etely redesigning the clamp to remove the long flat area that encouraged the break due to leverage. In the past six

PAGE 113

113 inch lengths were implemented due to reports that the smaller two inch stabilizer clamps were kinking the messenger cable during back and fo rth movement. Two span wire clamps were used during testing, each from a different manufacturer. The span wire clamps were of similar construction; however, the set from one manufacturer had a small piece which consistently broke during testing ( Figure 6 1 ). The broken piece was noticeably thinner than the equivalent piece from the other manufacturer. Either the thickness or the material should be changed on this part in order to increase strength.

PAGE 114

114 Figure 6 1 Failure of the joint between the top signal head and two way bracket (Test 22) (Photo courtesy of author) Figure 6 2 Worn serrated teeth on two way bracket which connects to top signal head (Test 22) (Photo cou rtesy of author)

PAGE 115

115 Figure 6 3 Failure at connection for the backplate on polycarbonate signal (Test 22) (Photo courtesy of author) Figure 6 4 Failed backplate material at connection point ( Test 27) (Photo courtesy of author)

PAGE 116

116 Figure 6 5 Failed sand cast aluminum stabilizer clamp : A) parts shown lain separated; B) parts shown together (Photos courtesy of author) A B

PAGE 117

117 Figure 6 6 Span wire clamp with broken piece (Photo courtesy of author)

PAGE 118

118 CHAPTER 7 CONCLUSIONS, SUMMARY AND RECOMMENDATION S 7.1 Summary A total of 33 tests were performed to compare the performance of dual cable systems with pipe hangers, pivotal hangers, and cabl e hangers and the single cable system with a direct connection. The orientation was varied for the traffic signal and the Hurricane Simulator. The material type and number of the signals was also varied. Data on wind speed, signal rotations, catenary and m essenger cable tensions, and cable displacements was collected. The rotations and cable tensions were compared to evaluate the behavior of each system. 7.2 Conclusions Test results showed that the dual cable system with pivotal or cable hangers experienced similar rotations to the single cable system. However, the single cable system allowed the signal to continuously swing when exposed to high wind loads. The pipe hanger underwent much lower rotations. When an additional signal was added to the span, there was no significant change to the rotations in the dual cable systems. The second signal on the single cable system caused the signals to swing more than with one signal. Using polycarbonate signals caused a slight increase in rotations when wind speeds we re between 40 mph and 100 mph. The use of a polycarbonate signal on the single cable system caused very erratic swinging and failure in some cases. The discontinuous messenger system experienced less rotation at high wind speeds than the single cable syste m and the dual cable system with pivotal and cable hangers. The discontinuous messenger did not sway even though the messenger cable was not whole.

PAGE 119

119 For all dual cable support systems, the catenary cables generally had decreasing tension with increased win d loads, while the messenger cables experienced significant increases in tension with increased wind loads. Messenger cable tension was lower in most cases with the use of pivotal or cable hangers instead of the pipe hanger. The single cable system experie nced very little change in the average cable tension, but there were fluctuations in the cable tension while the signal was swinging at higher wind loads. With the addition of a second signal there was very little effect to the cable tensions of the dual c able system, but the single cable system experienced high peaks in tension from the signals swinging. The polycarbonate signal had minimal effects on the tension of dual cable systems; however, the single cable system experienced high peaks in tension from the erratic swinging during higher wind speeds. The peak tension for the single cable system with a polycarbonate signal exceeded the peak tension of the single cable system with two aluminum signals. For all cases, the catenary tension of the discontinu ous messenger cable system was slightly lower than the single cable system and the dual cable system with the pipe hanger, and higher than the dual cable system with the pivotal and cable hangers. Cable translation measurements indicated that the pipe han ger forced the catenary cable forward when the signal rotated backwards. The catenary cable for the pivotal and cable hangers blew backwards after the signal weight and wind forces were transferred to the messenger cables. The displacement of the messenger cables significantly exceeded the displacement of the catenary cables due to strains imposed on the messenger cables from the wind load. The single cable system experienced

PAGE 120

120 cable displacements that were much higher than for either cable in the dual cable system due to the single cable system simply moving in the wind as a pendulum. The messenger cable force increase with wind loading on the dual cable system must be considered for strain pole design. Based on the data collected, a force coefficient of 0.4 5 is recommended for use when designing dual cable systems that restrict signal rotation, and a force coefficient of 0.2 should be used when designing dual cable systems that allow rotation of the signal. These are the maximum values that were calculated, and provided a conservative value for calculating design wind forces. The force coefficient should be used to determine a lateral wind force acting on the messenger cable Future research should focus on localized stresses since this project as well as pre vious projects show strain pole designs are conservative and failures occur within the components of the traffic signal and support system. 7.3 Recommendations Based on comments and suggestions following testing, several recommendations can be made for f uture research. Many limitations met during testing could be addressed with the use of wind tunnel testing. The use of a wind tunnel would eliminate many restrictions and allow for longer spans and more signals on a span. The tests for this project were un able to cause the dynamic behavior experienced in hurricanes. The damage to traffic signals observed after hurricanes suggests that many signals flip over the span wires. There was no evidence of this behavior during the full scale tests performed for this project. In future related projects, a few additional variables and combinations should be included to better understand the behavior of the various cable support systems. The

PAGE 121

121 discontinuous messenger cable system was a minor consideration during this pro ject. The results show that it may provide an effective remedy to the high messenger cable and strain pole stresses experienced in dual cable systems. The discontinuous messenger cable system required testing with multiple signals; however the use of multi ple signals provides more realistic results. This system should be tested at more orientations and with different types of hangers to better understand if it would provide the necessary hurricane resistance to prevent the extensive damage experienced by ex isting dual cable systems. The results from using polycarbonate signals with the single cable system show that it is more prone to damage and failure. The data for tests with backplate failure show that without the backplate, the signal behaved more like the aluminum signal with slightly higher rotations. The behavior of both polycarbonate and aluminum signals with and without backplates should be considered to determine if the use of backplates should be reduced. Discussions with Dan Weisburg, P.E., Dire ctor, and Scott Philbrick of the Traffic Division of Engineering and Public Works in Palm Beach County revealed that the cable hanger support system has been used for several decades in Palm Beach County. In the hurricanes of 2004, the predominant traffic signal failures in Palm Beach County occurred in either the bottom of the disconnect box or top of the signals and not in the cable hangers. This indicated that the cable hanger support system significantly reduced failures associated with the hanger (i. e., as opposed to the adjustable strap hanger system). For the cable hanger system and others that reduce hanger failure the next weakest link in the system appears to be the connections at the bottom of the

PAGE 122

122 disconnect box and the top of the signals. Futu re research should add focus on eliminating these weak points.

PAGE 123

123 APPENDIX A SIGNAL ROTATION VS. WIND VELOCITY GRAPHS Figure A 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) (Hanger failu re at 115 mph) Figure A 2 Test 2: Pivotal Hanger with 90 Signal and 90 Simulator Orientations (Linear Load)

PAGE 124

124 Figure A 3 Test 3: Direct Connection with 90 Signal and 90 Simulator Orienta tions (Linear Load) Figure A 4 Test 4: Pipe Hanger with 90 Signal and 90 Simulator Orientations

PAGE 125

125 Figure A 5 Test 5: Pivotal Hanger with 90 Signal and 90 Simulator Orientations Figur e A 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator Orientations

PAGE 126

126 Figure A 7 Test 7: Direct Connection with 90 Signal and 90 Simulator Orientations Figure A 8 Test 8: Direct Connection with 90 Signal and 90 Simulator Orientations (Polycarbonate Signal) (Backplate failure at 107 mph)

PAGE 127

127 Figure A 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations Fi gure A 10 Test 10: Pipe Hanger with 45 Signal and 45 Simulator Orientations Figure A 11 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations

PAGE 128

128 Figure A 12 Test 12: Pivotal Hanger with 45 Signal and 45 Simulator Orientations Figure A 13 Test 13: Cable Hanger with 90 Signal and 45 Simulator Orientations

PAGE 129

129 Figure A 14 Test 14: Cable Hanger with 45 Signal and 45 Simulator Orientations Figure A 15 Test 15: Direct Connection with 90 Signal and 45 Simulator Orientations Figure A 16 Test 16: Direct Connection wit h 45 Signal and 45 Simulator Orientations

PAGE 130

130 Figure A 17 Test 17: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) Figure A 18 Test 18: Pivotal Hanger with 90 Signa l and 45 Simulator Orientations (Two Signals)

PAGE 131

131 Figure A 19 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) Figure A 20 Test 20: Discontinuous Messenger w ith 90 Signal and 45 Simulator Orientations (Two Signals)

PAGE 132

132 Figure A 21 Test 21: Direct Connection with 90 Signal and 45 Simulator Orientations (Two Signals) Figure A 22 Test 22: Direct Connection with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) (Signal Head Failure at 118 mph)

PAGE 133

133 Figure A 23 Test 23: Adjustable Strap Hanger with 90 Signal and 45 Simulator Orientations Figure A 24 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front)

PAGE 134

134 Figure A 25 Test 25: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate S ignal) Figure A 26 Test 26: Cable Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal)

PAGE 135

135 Figure A 27 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientatio ns (Messenger on Front) (Backplate failure at 115 mph) Figure A 28 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations

PAGE 136

136 Figure A 29 Test 29: Pivotal Hanger with 90 Signal a nd 12 Simulator Orientations Figure A 30 Test 30: Cable Hanger with 90 Signal and 12 Simulator Orientations

PAGE 137

137 Figure A 31 Test 31: Pivotal Hanger with 90 Signal and 12 Simulator Orienta tions (Polycarbonate Signals) Figure A 32 Test 32: Direct Connection with 90 Signal and 12 Simulator Orientations (Polycarbonate Signals)

PAGE 138

138 Figure A 33 Test 33: Direct Connection with 90 Signal and 12 Simulator Orientations

PAGE 139

139 APPENDIX B CABLE TENSIONS VS. W IND VELOCITY GRAPHS Figure B 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientations (Linear Load) (Hanger failure at 115 mph) Figure B 2 Test 2: Pivotal Hanger with 90 Signal and 90 Simulator Orientations (Linear Load)

PAGE 140

140 Figure B 3 Test 3: Direct Connection with 90 Signal and 90 Simulator Orientations (Linear Load) Fig ure B 4 Test 4: Pipe Hanger with 90 Signal and 90 Simulator Orientations

PAGE 141

141 Figure B 5 Test 5: Pivotal Hanger with 90 Signal and 90 Simulator Orientations Figure B 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator Orientations Figure B 7 Test 7: Direct Connection with 90 Signal and 90 Simulator Orientations

PAGE 142

142 Figure B 8 Test 8: Direct C onnection with 90 Signal and 90 Simulator Orientations (Polycarbonate Signal) (Backplate failure at 107 mph) Figure B 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations

PAGE 143

143 Figure B 10 Test 10: Pipe Hanger with 45 Signal and 45 Simulator Orientations Figure B 11 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations Figure B 12 Test 12: Piv otal Hanger with 45 Signal and 45 Simulator Orientations

PAGE 144

144 Figure B 13 Test 13: Cable Hanger with 90 Signal and 45 Simulator Orientations Figure B 14 Test 14: Cable Hanger with 45 Sign al and 45 Simulator Orientations

PAGE 145

145 Figure B 15 Test 15: Direct Connection with 90 Signal and 45 Simulator Orientations Figure B 16 Test 16: Direct Connection with 45 Signal and 45 Simul ator Orientations

PAGE 146

146 Figure B 17 Test 17: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) Figure B 18 Test 18: Pivotal Hanger with 90 Signal and 45 Simulator Orient ations (Two Signals)

PAGE 147

147 Figure B 19 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) Figure B 20 Test 20: Discontinuous Messenger with 90 Signal and 45 Sim ulator Orientations (Two Signals)

PAGE 148

148 Figure B 21 Test 21: Direct Connection with 90 Signal and 45 Simulator Orientations (Two Signals) Figure B 22 Test 22: Direct Connection with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) (Signal Head Failure at 118 mph)

PAGE 149

149 Figure B 23 Test 23: Adjustable Strap Hanger with 90 Signal and 45 Simulator Orientations Figure B 24 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front)

PAGE 150

150 Figure B 25 Test 25: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal) Figure B 26 Test 26: Cable Hanger with 90 Signal and 45 Simulator Orientations (Polycarbonate Signal)

PAGE 151

151 Figure B 27 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Messenger on Front) (Ba ckplate failure at 115 mph) Figure B 28 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations

PAGE 152

152 Figure B 29 Test 29: Pivotal Hanger with 90 Signal and 12 Simulator Orientatio ns Figure B 30 Test 30: Cable Hanger with 90 Signal and 12 Simulator Orientations

PAGE 153

153 Figure B 31 Test 31: Pivotal Hanger with 90 Signal and 12 Simulator Orientations (Polycarbonate Signal s) Figure B 32 Test 32: Direct Connection with 90 Signal and 12 Simulator Orientations (Polycarbonate Signals)

PAGE 154

154 Figure B 33 Test 33: Direct Connection with 90 Signal and 12 Simulator Or ientations

PAGE 155

155 APPENDIX C CABLE DISPLACEMENT V S. WIND VELOCITY GRA PHS Due to errors with instrumentation, there are no displacement data sets for the following tests: Tests 10, 13, 20, 21, 22, 25, 26, 30, and 32. For Test 24 there is only displacement data f or the messenger cable. Figure C 1 Test 1: Pipe Hanger with 90 Signal and 90 Simulator Orientations (Linear Load)(Hanger failure at 115 mph) Figure C 2 Test 2: Pivotal Hanger with 90 Sig nal and 90 Simulator Orientations (Linear Load)

PAGE 156

156 Figure C 3 Test 3: Direct Connection with 90 Signal and 90 Simulator Orientations (Linear Load) Figure C 4 Test 4: Pipe Hanger with 90 S ignal and 90 Simulator Orientations

PAGE 157

157 Figure C 5 Test 5: Pivotal Hanger with 90 Signal and 90 Simulator Orientations Figure C 6 Test 6: Cable Hanger with 90 Signal and 90 Simulator Orie ntations

PAGE 158

158 Figure C 7 Test 7: Direct Connection with 90 Signal and 90 Simulator Orientations Figure C 8 Test 8: Direct Connection with 90 Signal and 90 Simulator Orientations (Polycarbona te Signal) (Backplate failure at 107 mph) Figure C 9 Test 9: Pipe Hanger with 90 Signal and 45 Simulator Orientations

PAGE 159

159 Figure C 10 Test 11: Pivotal Hanger with 90 Signal and 45 Simulator Orientations Figure C 11 Test 12: Pivotal Hanger with 45 Signal and 45 Simulator Orientations

PAGE 160

160 Figure C 12 Test 14: Cable Hanger with 45 Signal and 45 Simulator Orientations Figure C 13 Test 15: Direct Connection with 90 Signal and 45 Simulator Orientations

PAGE 161

161 Figure C 14 Test 16: Direct Connection with 45 Signal and 45 Simulator Orientations Figure C 15 Test 17: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Two Signals)

PAGE 162

162 Figure C 16 Test 18: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Two Signals) Figure C 17 Test 19: Cable Hanger with 90 Signal and 45 Simulator Orientations (Two Signals)

PAGE 163

163 Figure C 18 Test 23: Adjustable Strap Hanger with 90 Signal and 45 Simulator Orientations Figure C 19 Test 24: Pivotal Hanger with 90 Signal and 45 Simulator Orientations (Messenger clamped to front of hanger)

PAGE 164

164 Figure C 20 Test 27: Pipe Hanger with 90 Signal and 45 Simulator Orientations (Messenger clamped to front of hanger) (Failure of Backplate at 115 mph) Figure C 21 Test 28: Pipe Hanger with 90 Signal and 12 Simulator Orientations

PAGE 165

165 Figure C 22 Test 30: Cable Hanger with 90 Si gnal and 12 Simulator Orientations Figure C 23 Test 31: Pivotal Hanger with 90 Signal and 12 Simulator Orientations (Polycarbonate Signal)

PAGE 166

166 Figure C 24 Test 33: Direct Connection with 90 Signal and 12 Simulator Orientations

PAGE 167

167 APPENDIX D EQUIPMENT CATALOG Part: Pipe Manufacturer: Various Description: Schedule 40 aluminum; Weight: 2.1 lbs Test Used: 1 Part: Pi pe Manufacturer: Various Description: Schedule 80 aluminum; Weight: 1.3 lbs Tests Used: 4, 9, 10, 17, 27, 28 Part: EC 2079 & 2079 B Manufacturer: Engineered Castings Description: Sp an wire hanger and hardware Weight: 1.3 lbs Tests Used: 6, 13, 14, 19, 26, 30 Part: EC 2079 S & 2079 C Manufacturer: Engineered Castings Description: Span wire hanger and hardware; modified design for cable hanger Weight: 1.4 lbs Te sts Used: 6, 13, 14, 19, 26, 30

PAGE 168

168 Part: 1157 BT Manufacturer: Engineered Castings Description: Multi eye balancer Weight: 1.0 lb Tests Used: 3, 6, 7, 8, 13, 14, 15, 16, 19, 21, 22, 26, 30, 32, 33 Part: EC 2051 T3 Manufacturer: Engine ered Castings Description: Thread to tri stud adapter Weight: 0.7 lb Tests Used: 1, 4, 9, 10, 17, 27, 28 Part: 1900 Manufacturer: Cost Cast Description: Disconnect box Weight: 7.0 lbs Tests Used: All Part: 1906 Manufacturer: Cost Cast Description: Adjustable strap hanger (6" stabilizer clamp also shown in picture) Weight: 2.2 lbs Tests Used: 2, 5, 11, 12, 18, 20, 23, 24, 25, 29, 31 (Only top section was used with pivotal hanger tests)

PAGE 169

169 Part: 1976 Manufacturer: Cos t Cast Description: Wire entrance fitting Weight: 3.0 lbs Tests Used: 1, 4, 9, 10, 17, 27, 28 Part: 8880 Manufacturer: Cost Cast and Signal Safe Description: Pivotal assembly Weight: 3.5 lbs Tests Used: 2, 5, 11, 12, 18, 24, 25, 29, 31 stabilizer clamps Manufacturer: Cost Cast, Engineered Castings, and Pelco Description: Clamps m essenger cable to pivotal and strap hangers Weight: 0.7 lb Tests Used: 2, 5, 11, 12, 18, 20, 23, 24, 25, 29, 31 Part: Cable hanger Manuf acturer: N/A wire strand cable, Weight: 1.0 lb (Not including span wire hangers) Tests Used: 6, 13, 14, 19, 26, 30

PAGE 170

170 Part: Clamp Manufacturer: Various Description: Drop pipe cable clamp We ight: 0.8 lb Tests Used: 1, 4, 9, 10, 17, 27, 28 Part: Aluminum Signals assemblies Manufacturer: Signal head: McCain (provided by Control Technologies) Signal hardware: Cost Cast LED module: Leotek Description: 5 section cluster and 3 section as sembly with louvered backplates, tunnel visors, and LED signal modules Weight: 5 section : 73 lbs 3 section : 40 lbs Tests Used: 5 section : 1 7, 9 21, 23, 24, 27 30, 33; 3 section : 17 21 Part: Polycarbonate signal assemblies Manufacturer: Signal h ead: Econolite and McCain (provided by Control Technologies) Signal hardware: Cost Cast LED module: Dialight and Leotek Description: 5 section cluster with louvered backplate, tunnel visors, and LED signal modu les; top head constructed of aluminum and re maining heads constructed of polycarbonate Weight: 53.5 lbs Tests Used: 8, 22, 25, 26, 31, 32

PAGE 171

171 LIST OF REFERENCES Alampalli, S. l oads on u ntethered span wire t raffic s ignal p Special Report 126 New York State Department of Transpo rtation, Albany, NY. American Association of State Highway Transportation Officials (AASHTO). (2009). Standard Specifications for Structural Supports for Highway Signs, Luminaires and Traffic Signals, 5 th Ed., AASHTO, Washington, D.C. American Society of Civil Engineers (ASCE). (2010). ASCE 7 10: Minimum Design Loads for Buildings and Other Structures ASCE, Reston, VA. FDOT WPI No. 051073 1 University of Florida, Engineering and Industrial Experiment Station, Gainesville, FL. Cook, R.A., Hoit, M.I., Ashley, F.K. Christou, P.M., Drost, K., & Wajek, S.L. (1993). Tw FDOT WPI No. 0510653, Contract No. C 4430 University of Florida, Engineering and Industrial Experiment Station, Gainesville, FL. Supported Traffic S FDOT Contract No. BD545 57 University of Florida, Department of Civil and Coastal Engineering, Gainesville, FL. Program for Signal Pole and Span Wire Assembl ies with Two Point Connection System FDOT WPI No. 0510653, Contract No. B 8409 University of Florida, Engineering and Industrial Experiment Station, Gainesville, FL. n Program for Signal Pole and Span Wire Assemblies with Two Point Connection FDOT WPI No. 0510653, Contract No. B 9924 University of Florida, Engineering and Industrial Experiment Station, Gainesville, FL. Hoit, M.I., Cook, R.A., Wajek, S.L., & FDOT WPI No. 0510653, Contract No. C 8409 University of Florida, Engineering and Industrial Experiment Station, Gainesville, FL. Institute of Transportat Council Committee, (2005). Vehicle Traffic Control Signal Heads: Light Emitting Diode (LED Circular Signal Supplement, ITE, Washington, D.C.

PAGE 172

172 State of Florida Department of Transportation (FDOT). Tallahassee, FL State of Florida Department of Transportation (FDOT). (2007). Manual of Uniform Minimum Standards for Design, Construction and Maintenance f or Streets and Highways, FDOT, Tallahassee, FL.

PAGE 173

173 BIOGRAPHICAL SKETCH Je ssica L. Rigdon was born in Panama City, FL, to John N. Rigdon Jr. and Terry L. Rigdon. After graduating among the top ten of her class at A. Crawford Mosley High School in 2006, she a ttended Gulf Coast Community College and was awarded an associate of the arts degree in engineering in July 2007. Jessica then attended the University of Florida where she was accepted into the Department of Civil and Coastal Engineering. She graduated wit engineering in May 2010. As an undergraduate student Jessica was a participant and officer in the UF ASCE Student Chapter and Concrete Canoe Team. For the last year of her undergraduate degree, she was a student assistant for multiple honors and graduate students doing research in the field of structural engineering. Jessica began her master of engineering degree and started research with Dr. Cook and Dr. Masters in August 2010, and she completed her degree in December 2011.