sponsored - Auburn University

EVALUATION OF FATIGUE CRACKING IN 1-65 MOBILE DELTA CROSSING BRIDGES

Volume IV: Executive Summary Project Number ST 2019-16

by

J. Michael Stallings Thomas E. Cousins

sponsored by

The State of Alabama Highway Department Montgomery, Alabama

December 1993

ACKNOWLEDGEMENT

The material contained herein was obtained in connection with a research project,

"Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges" (ST-2019-16),

conducted by the Highway Research Center at Auburn University. The research project was

sponsored by the State of Alabama Highway Department. The traffic control, test load vehicles,

bridge inspection vehicles and operators were provided by the Alabama Highway Department's

Ninth Division and Maintenance Bureau. The interest, cooperation, and assistance of many

personnel from the Alabama Highway Department is gratefully acknowledged. The assistance of

graduate students Bradley P. Christopher, Charles B. Reid, Richard K. Rotto, and Scott K.

Rutland during the instrumentation. field testing. data reduction, and report preparation portions of

this project is also gratefully acknowledged.

DISCLAIMER

The contents of this report reflect the views of the authors who are responsible for the

facts and accuracy of the data presented herein. The contents do not necessarily reflect the

o'fficial views or policies of the State of Alabama Highway Department or Auburn University. The

report does not constitute a standard, specification, or regulation.

~~----...... --.. ----- --....... ---~

ABSTRACT

The 1-65 Mobile Delta Crossing Bridges constitute a six mile section (northbound and

southbound) of interstate highway over the Mobile River delta plains located approximately 20

miles north of Mobile, Alabama. The bridges consist of precast concrete spans, steel plate

girder spans, and twin steel tied arch spans. In 1990 and 1991, Alabama Highway

Department bridge inspectors discovered fatigue cracks at connections of floorbeams and

floortrusses to the plate girders. The fatigue cracks were found to be the result of out-of-plane

distortion of the girder web. Field measurements of the fatigue stresses at the connections

were made before and after various repairs were perform to evaluate the effectiveness of the

repairs. Bridge inspectors also found cracked (failed) welds at filler plates between the

double angles of the floortruss members along the tied arch spans. These cracks were found

to result from impacted rust between the filler plates and the angles .. Field measurements of

the traffic loading stresses were made, and the potential for future fatigue cracking at the filler

plate welds was evaluated. Recommendations for repair and maintenance of the bridges

were made.

This volume is the fourth in a series. The other volumes in the series are:

EVALUATION OF FATIGUE CRACKING IN 1-65 MOBILE DELTA CROSSING BRIDGES, Volume I: Floorbeam-Girder Connections

EVALUATION OF FATIGUE CRACKING IN 1-65 MOBILE DELTA CROSSING BRIDGES, Volume II: Filler Plate Weld Cracks

EVALUATION OF FATIGUE CRACKING IN 1-65 MOBILE DELTA CROSSING BRIDGES, Volume III: Floortruss-Girder Connections

Volume I: Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges - Floorbeam-Girder Connections

TABLE OF CONTENTS

Chapter 1 INTRODUCTION BACKGROUND BRIDGE DESCRIPTION LOCATIONS AND EXTENT OF CRACKING RIGID AlTACHMENT REPAIR PROJECT OBJECTIVES METHODOLOGY AND SCOPE OF REPORT

Chapter 2 LITERATURE REVIEW 12 FATIGUE CRACKING AT FLOORBEAM CONNECTION PLATES LABORATORY TESTS INSTRUMENTATION CYCLE COUNTING METHODS EFFECTIVE STRESS RANGE

Chapter 3 TEST LOCATIONS 33

Chapter 4 INSTRUMENTATION AND DATA ACQUISITION 42 WORKING ENVIRONMENT INSTRUMENTATION DATA ACQUISITION

Chapter 5 DATA COLLECTION AND REDUCTION 62 CALIBRATION TESTS RANDOM TRUCK TESTS DATA REDUCTION

Chapter 6 RESULTS AT PIER FLOORBEAM CONNECTIONS 77 BEHAVIOR AT TEST LOCATIONS WEB GAP STRESSES FROM STATIC TESTS EFFECTIVE STRESS RANGES FROM TRUCK TRAFFIC

Chapter 7 EVALUATION OF CRACKING AND REPAIR METHODS 104 CAUSE OF FATIGUE CRACK INITIATION ASSESSMENT OF HOLE DRILLING AS A REPAIR METHOD EFFECTIVENESS OF THE POSITIVE AlTACHMENT RETROFIT

Chapter 8 RESULTS AT FLOORBEAM CONNECTIONS AWAY FROM PIERS 132 RESULTS AT FLOORBEAM TWO RESULTS AT FLOORBEAM THREE

ii


TABLE OF CONTENTS (continued)

Chapter 9 CONCLUSIONS AND RECOMMENDATIONS SUMMARY AND CONCLUSIONS RECOMMENDATIONS

REFERENCES

APPENDICES A. S7-5 TABULATED RANDOM TRUCK HISTOGRAM DATA B. S7-9 TABULATED RANDOM TRUCK HISTOGRAM DATA C. S5-2 TABULATED RANDOM TRUCK HISTOGRAM DATA D. S7-12 TABULATED RANDOM TRUCK HISTOGRAM DATA

iii

151

155

158

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure 18.

Figure 19.


LIST OF FIGURES

Floorbeam-Girder Detail at a Girder Support

Illustration of Out-of-Plane Behavior in Web Gap Region (from Fisher et al. 1990)

Typical Crack Found at Floorbeam Connection Plate Details

Positive Attachment Retrofit Detail

Positive Attachment Retrofit Detail (Section B from Figure 4)

Out-of-Plane Bending Test Setup (from Fisher et al. 1979b)

Test Setup for Distortion-Induced Web Gap Fatigue Cracking (from Keating and Fisher 1987)

History of Web Gap Fatigue Cracking (from Keating and Fisher 1987)

Out-of-Plane Bending Test Setup for Third Series of Tests (from Fisher et al. 1990)

Defined Crack Length, ar (from Fisher et al. 1990)

Typical Web Gap Region Gage Arrangements

Illustration of Peak-to-Peak Method

Four Test Locations on Mobile Delta Crossing Bridges

Position of Test Piers on the Continuous Spans

Cross-Section of Bridges Illustrating Transverse Floorbeams

Location of Web Gap Regions on Cross-Section

Typical Web Gap Region: (a) Elevation of Inside Face of Girder; (b) Section View

Work Platform

Reach-all Truck

iv

EAG.E

4

5

6

8

9

15

18

19

20

23

26

30

34

35

37

37

41

43

43

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Figure 29.

Figure 30.

Figure 31.

Figure 32.

Figure 33.

Figure 34.

Figure 35.


LIST OF FIGURES (continued)

Numbering and Locations of Strain Gages on Girder Web

Numbering and Modified Locations of Strain Gages on Girder Web

Numbering and Locations of In-Plane Bending Gages on Girder

Numbering and Locations of Strain Gages on Floorbeam and Stringers

LVDT Arrangement at Pier Floorbeam

LVDT Arrangement at First Floorbeam South of Pier

LVDT Arrangement at Second Floorbeam South of Pier

Environmental Chamber

Generator with Hangers Attached

Trucks Used for Calibration Tests: (a) Dump Truck; (b) Sand Spreader; (c) Lowboy

Test Lanes A, B, and C Used for Calibration Tests

Dump Truck Positions Over Floorbeams During Static Tests

Lowboy Positions Over Floorbeams During Static Tests

Typical Truck Crossing Record (400 Samples/sec)

Truck Crossing Shown in Figure 34 Compressed by Averaging Every 25 Scans

Bottom Flange Strain Measured on the Outside Girder at Pier (a) S7-5 for Lowboy in the Outside Lane at 60 mph, (b) S7-5 for Dump Truck in the Outside Lane at 58 mph, (c) S5-2 for Lowboy in the Outside Lane at 65 mph, (d) S5-2 for Dump Truck in the Outside Lane at 60 mph

v

46

46

48

48

56

57

57

60

60

67

67

70

70

75

75

79

Figure 36.

Figure 37.

Figure 38.

Figure 39.

Figure 40.

Figure 41.

Figure 42.

Figure 43.

Figure 44.

Figure 45.

Figure 46.

Figure 47.

Figure 48.



Web Gap Displacement Measured on the Inside Girder at Pier (a) S7-5 for Lowboy in the Outside Lane at 60 mph, (b) 57-5 for Dump Truck in the Outside Lane at 58 mph, (c) S5-2 for Lowboy in the Outside Lane at 65 mph, (d) S5-2 for Dump Truck in the Outside Lane at 60 mph

Response of Gages IN 1 and IN2 on the Inside Girder at Pier S7-5 During Lowboy Crawl Pass in the Outside Lane

Response of Gages IN 1 and IN2 on the Inside Girder at Pier S7-5 During Lowboy Fast Pass in the Outside Lane

Response of Gages IN 1 and IN2 on the Inside Girder at Pier S7-9 During Lowboy Crawl Pass in the Outside Lane

Response of Gages IN 1 and IN2 on the Inside Girder at Pier S7-9 During Lowboy Fast Pass in the Outside Lane

Response of Gages ON1 and IS2 on the Inside Girder at Pier S7-12 During Lowboy Crawl Pass in the Outside Lane

Response of Gages ON1 and IS2 on the Inside Girder at Pier S7-12 During Lowboy Fast Pass in the Outside Lane

Static Stress Distribution on Outside Girder at Pier S7-5 (84 in. Girder) Before Retrofit with Dump Truck in Outside Lane (Worst Case)

Static Stress Distribution on Inside Girder at Pier S7-5 (84 in. Girder) Before Retrofit with Dump Truck in Outside Lane (Worst Case)

Static Stress Distribution on Outside Girder at Pier S5-2 (84 in. Girder) Before Retrofit with Dump Truck in Outside Lane (Worst Case)

Static Stress Distribution on Inside Girder at Pier S7-12 (96 in. Girder) Before Retrofit with Dump Truck in Outside Lane (Worst Case)

Static Stress Distribution on Outside Girder Pier S7-9 (96 in. Girder) Before Retrofit with Lowboy in Outside Lane (Worst Case)

Static Stress Distribution on Inside Girder of Pier S7-9 (96 in. Girder) Before Retrofit with Lowboy in Outside Lane (Worst Case)

vi

81

84

84

85

85

86

86

88

88

89

89

90

90

Figure 49.

Figure 50.

Figure 51.

Figure 52.

Figure 53.

Figure 54.

Figure 55.

Figure 56.

Figure 57.

Figure 58.

Figure 59.

Figure 60.

Figure 61.

Figure 62.



Histogram of Extrapolated Random Data at the End of the Inside Connection Plate Weld on Outside Girder of Pier 87-5 Before Retrofit

Histogram of Extrapolated Random Data at the Inside Web-Flange Weld Toe on Outside Girder of Pier 87-5 Before Retrofit

Histogram of Extrapolated Random Data at the Outside Web-Flange Weld Toe on Outside Girder of Pier 87-5 Before Retrofit

Histogram of Extrapolated Random Data at the End of the Inside Connection Plate Weld on Inside Girder of Pier 87-5 Before Retrofit

Histogram of Extrapolated Random Data at the Inside Web-Flange Weld Toe on Inside Girder of Pier 87-5 Before Retrofit

Histogram of Extrapolated Random Data at the Outside Web-Flange Weld Toe on Inside Girder of Pier 87-5 Before Retrofit


Histogram of Extrapolated Random Data at the Inside Web-Flange Weld Toe on Outside Girder of Pier 85-2 Before Retrofit


Histogram of Extrapolated Random Truck at the Inside Web-Flange Weld Toe on Outside Girder of Pier 87-9 Before Retrofit

Histogram of Extrapolated Random Truck at the Outside Web-Flange Weld Toe on Outside Girder of Pier 87-9 Before Retrofit

Histogram of Extrapolated Random Data at the End of the Inside Connection Plate Weld on Inside Girder of Pier 87-9 Before Retrofit

Histogram of Extrapolated Random Truck at the Inside Web-Flange Weld Toe on Inside Girder of Pier 87-9 Before Retrofit

Histogram of Extrapolated Random Truck at the Outside Web-Flange Weld Toe on Inside Girder of Pier 87-9 Before Retrofit

vii

93

93

94

94

95

95

96

96

97

97

98

98

99

99

Figure 63.

Figure 64.

Figure 65.

Figure 66.

Figure 67.

Figure 68.

Figure 69.

Figure 70.

Figure 71.

Figure 72.

Figure 73.

Figure 74.

Figure 75.



Histogram of Extrapolated Random Data at the End of the Inside Connection Plate Weld on Inside Girder of Pier S7 -12 Before Retrofit 100

Histogram of Extrapolated Random Truck at the Inside Web-Flange Weld Toe on Inside Girder of Pier S5-2 Before Retrofit 100

Fatigue Data Reported by Fisher et al. (1990) Illustrating the Fatigue Resistance of Connection Plate Details Subjected to Out-of-Plane Distortions 105

Static Stress Distribution on Outside Girder at Pier S7 -5 (84 in. Girder) After Retrofit with Dump Truck in Outside Lane (Worst Case) 110

Static Stress Distribution on Inside Girder at Pier S7-5 (84 in. Girder) After Retrofit with Dump Truck in Outside Lane (Worst Case) 110

Static Stress Distribution on Outside Girder at Pier S7-9 (96 in. Girder) After Retrofit with Dump Truck in Outside Lane (Worst Case) 111

Static Stress Distribution on Inside Girder of Pier S7-9 (96 in. Girder) After Retrofit with Dump Truck in Outside Lane (Worst Case) 111

Before and After Retrofit Strain Responses for Gage OS1 on the Outside Girder of Pier S7-5 for the Lowboy Fast Run in the Outside Lane 113

Before and After Retrofit Strain Responses for Gage IN1 on the Inside Girder of Pier S7-5 for the Lowboy Fast Run in the Inside Lane 113

Before and After Retrofit Strain Responses for Gage IN1 on the Outside Girder of Pier S7-9 for the Lowboy Fast Run in the Outside Lane 114

Before and After Retrofit Strain Responses for Gage IN 1 on the Inside Girder of Pier S7-9 for the Lowboy Fast Run in the Outside Lane 114

Histogram of Out-of-Plane Displacement of the Outside Girder Web Gap at Pier S7-5 Before Retrofit 119

Histogram of Out-of-Plane Displacement of the Outside Girder Web Gap at Pier S7 -5 After Retrofit 119

viii



Figure 76. Histogram of Out-of-Plane Displacement of the Inside Girder Web Gap at Pier 87-5 Before Retrofit

Figure 77. Histogram of Out-of-Plane Displacement of the Inside Girder Web Gap at Pier 87-5 After Retrofit

Figure 78. Histogram of Out-of-Plane Displacement of the Inside Girder Web Gap at Pier 87-9 Before Retrofit

Figure 79. Histogram of Out-of-Plane Displacement of the Inside Girder Web Gap at Pier 87 -9 After Retrofit

Figure 80. Histogram of Out-of-Plane Displacement of the Outside Girder Web Gap at Pier 87-9 Before Retrofit

Figure 81. Histogram of the Random Data for Gage 081 on the Outside Girder at 87-5 Before Retrofit

Figure 82. Histogram of the Random Data for Gage 081 on the Outside Girder at 87-5 After Retrofit

Figure 83. Histogram of the Random Data for Gage 182 on ,the Outside Girder at 87-5 Before Retrofit

Figure 84. Histogram of the Random Data for Gage 182 on the Outside Girder at 87-5 After Retrofit

Figure 85. Histogram of the Random Data for Gage 081 on the Inside Girder at 87-5 Before Retrofit

Figure 86. Histogram of the Random Data for Gage 081 on the Inside Girder at 87-5 After Retrofit

Figure 87. Histogram of the Random Data for Gage 182 on the Inside Girder at 87-5 Before Retrofit

Figure 88. Histogram of the Random Data for Gage 182 on the Inside Girder at 87-5 After Retrofit

Figure 89. Histogram of the Random Data for Gage 081 on the Outside Girder at 87-9 Before Retrofit

ix

120

120

121

121

122

122

123

123

124

124

125

125

126

126



Figure 90. Histogram of the Random Data for Gage OS1 on the Outside Girder at S7 -9 After Retrofit 127

Figure 91. Histogram of the Random Data for Gage IS2 on the Outside Girder at S7 -9 Before Retrofit 127

Figure 92. Histogram of the Random Data for Gage IS2 on the Outside Girder at S7 -9 After Retrofit 128

Figure 93. Histogram of the Random Data for Gage OS1 on the Inside Girder at S7 -9 Before Retrofit 128

Figure 94. Histogram of the Random Data for Gage OS1 on the Inside Girder at S7 -9 After Retrofit 129

Figure 95. Histogram of the Random Data for Gage IS2 on the Inside Girder at S7 -9 Before Retrofit 129

Figure 96. Histogram of the Random Data for Gage IS2 on the Inside Girder at S7 -9 After Retrofit 130

Figure 97. Static Stress Distribution for Outside Girder at Pier S5-2 Floorbeam Two with Dump Truck in Outside Lane (Worst Case) 133

Figure 98. Static Stress Distribution for Inside Girder at Pier S7 -12 Floorbeam Two with Dump Truck in Outside Lane (Worst Case) 133

Figure 99. Histogram of Extrapolated Random Data at the Outside Web-Flange Weld on the Outside Girder at Pier S5-2 Floorbeam Two 134

Figure 100. Histogram of Extrapolated Random Data at the Outside Web-Flange Weld Toe on the Inside Girder at Pier S7-12 Floorbeam Two 134

Figure 101. Histogram of Extrapolated Random Data at the Location of the Floorbeam Two Connection Plate Weld on the Outside Face of the Girder at S5-2 141

Figure 102. Histogram of Extrapolated Random Data at the Location of Floorbeam Two Connection Plate Weld on the Outside Face of the Girder at S7-12 141

x



Figure 103. Histogram ot the Out-ot-Plane Displacements Recorded at Pier S5-2 Floorbeam Two 142

Figure 104. Histogram ot the Out-ot-Plane Displacements Recorded at Pier S7 -12 Floorbeam Two 142

Figure 105. Static Stress Distribution tor Outside Girder at Pier S5-2 Floorbeam Three with Lowboy in Outside Lane 148

Figure 106. Static Stress Distribution tor Inside Girder at Pier S7 -12 Floorbeam Three with the Lowboy in the Outside Lane 148

Figure 107. Histogram ot Out-ot-Plane Displacements at Pier S5-2 Floorbeam Three 149

Figure 108. Histogram ot the Out-ot-Plane Displacements at Pier S7-12 Floorbeam Three 149

xi

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

Table 19.

Table 20.


LIST OF TAB LES

Mobile Delta Crossing Bridges' Traffic History

Cycle Counting Methods

Web Gap Geometric Properties

Plate Girder Dimensions

Gages Installed at 87-5 Before Retrofit

Gages Installed at 87-9 Before Retrofit

Gages Installed at 85-2

Gages Installed at 87-12

Gages Installed at 87-5 After Retrofit

Gages Installed at 87-9 After Retrofit

Calibration Truck Weights for 87-5 Before Retrofit




Calibration Truck Weights for 87-5 After Retrofit

Calibration Truck Weights for 87-9 After Retrofit

Typical Event 8equence for Calibration Tests

Number of Random Truck Crossings Recorded at Each Test Location

Effective 8tress Ranges at Critical Locations from Random Truck Data Before Retrofit

8tress Ranges at Critical Locations Before Retrofit for Lowboy Test Truck Fast Runs

xii

2

28

39

40

49

50

51

52

53

54

65

65

65

66

66

66

69

72

92

102

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.


LIST OF TABLES (continued)

Static Calibration Test Results Before and After Positive Attachment

Average and Standard Deviation of Out-of-Plane Displacements for Random Data at S7 -5 Before and After the Positive Attachment 'Retrofit

Effective Stress Ranges at Gages OS1 and IS2 for Random Data Before and After Retrofit

Static Calibration Test Results at Floorbeam Two for Test Trucks over Floorbeam Two

Stress Ranges at Critical Locations for Lowboy Test Truck

Effective Stress Ranges at Critical Locations at Floorbeam Two

Static Calibration Test Results at Floorbeam Three for Test Trucks over Floorbeam Two

Static Calibration Test Results at Floorbeam Three for Test Trucks over Floorbeam Three

xiii

PAGE

115

116

118

136

137

140

145

146

Volume II: Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges - Filler Plate Weld Cracks

TABLE OF CONTENTS

Chapter 1 INTRODUCTION BACKGROUND PROJECT OBJECTIVES METHODOLOGY AND SCOPE OF REPORT

Chapter 2 BRIDGE DESCRIPTION AND TEST LOCATIONS TEST LOCATIONS

Chapter 3 INSTRUMENTATION AND DATA ACQUISITION WORKING ENVIRONMENT INSTRUMENTATION DATA ACQU ISITION

Chapter 4 DATA COLLECTION AND REDUCTION CALIBRATION TESTS RANDOM TRUCK TESTS DATA REDUCTION EFFECTIVE STRESS RANGE

Chapter 5 TEST RESULTS AND ANALYSES STATIC TEST RESULTS DYNAMIC TEST RESULTS COMPARISON OF RESULTS FROM FLOORTRUSSES 4,

7, AND 8 COMPARISON OF AXIAL AND EXTREME FIBER STRESS

RANGES SIGNIFICANCE OF LANE POSITION COMPARISON OF RESULTS FROM FIELD TESTS AND

STRUCTURAL ANALYSES

Chapter 6 EVALUATION OF POTENTIAL FOR FATIGUE CRACKING. CRITICAL EXTREME FIBER LOCATIONS COMPARISONS WITH AASHTO FATIGUE LIMITS FRACTURE MECHANICS APPROACHES


REFERENCES

xiv

9

21

37

49

106

119

122

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

Figure 18.


LIST OF FIGURES

Typical Double Angle with Filler Plates

Filler Plate Weld Crack

Filler Plate Weld Crack and Solidified Rust

Fatigue Resistance of Category C Details (from Keating and Fisher 1986)

Typical Floortruss Section

Mobile Delta River Crossing Bridges

2

3

4

8

10

13

Elevation View of Southbound Arch Span (Looking from Northbound Lane) 14

Identification Scheme for Floortruss Members (Looking South)

Existing Filler Plate Weld Crack Locations (denoted by letters): (a) Floortruss 4; (b) Floortruss 8 (Looking South)

Plan View of Work Platforms at Floortruss Test Location

Gage Locations on Bottom Chord Members

Bottom Chord Gage Location Details (See Figure 11): (a) Detail A; (b) Section B-B

Instrumented Diagonal Members

Gage Locations on Diagonal Members

Gage Location Details for Members 02, 03, 04, and 05: (a) Detail C (See Figure 14); (b) Section C1-C1

Gage Location Details for Members 01 and 06: (a) Detail D (See Figure 14); (b) Section 01-01

Environmental Chamber

Plan View of Trucks Used for Calibration Tests: (a) 3-Axle; (b) 5-Axle

xv

17

18

22

25

26

28

29

30

31

35

39

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Figure 29.

Figure 30.

Figure 31.

Figure 32.

Figure 33.

Figure 34.

.Figure 35.

Figure 36.

Figure 37.

Figure 38.



Test Lanes A, B, and C Used for Calibration Tests

Plan View of Southbound Lanes with Static Calibration Truck Positions P1 through P7

Alignment of Test Trucks for Static Tests: (a) 3-Axle Truck; (b) 5-Axle Truck

Illustration of Peak to Peak Method

Axial Stress in Member 01 of Floortruss 4 with Test Trucks in Slow Lane


Axial Stress in Member 01 of Floortruss 4 with Test Trucks in Fast Lane













Axial Stress in Member'06 of Floortruss 8 with Test Trucks in Fast Lane

xvi

39

43

43

46

50

50

51

51

52

52

53

53

54

54

55

55

56

56

57

57

Figure 39.

Figure 40.

Figure 41.

Figure 42.

Figure 43.

Figure 44.

Figure 45.

Figure 46.

Figure 47.

Figure 48.

Figure 49.

Figure 50.

Figure 51.



Typical Axial Strain Records for Member 01 due to Test Truck Fast Runs in the Slow Lane: (a) 3-Axle at Floortruss 4; (b) 3-Axle at Floortruss 8; (c) 5-Axle at Floortruss 4; (d) 5-Axle at Floortruss 8

Typical Axial Strain Records for Member 02 due to Test Truck Fast Runs in the Slow Lane: (a) 3-Axle at Floortruss 4; (b) 3-Axle at Floortruss 8; (c) 5-Axle at Floortruss 4; (d) 5-Axle at Floortruss 8

Typical Axial Strain Records for Member 83 due to Test Truck Fast Runs in the Slow Lane: (a) 3-Axle at Floortruss 4; (b) 3-Axle at Floortruss 8 (c) 5-Axle at Floortruss 4; (d) 5-Axle at Floortruss 8

Axial Stress Ranges (ksi) in Floortruss 4 due to 5-Axle Test Truck: (a) Truck in Slow Lane at 53 mph; (b) Truck in Fast Lane at 53 mph




Static Axial Stress in Member 01 from 5-Axle Truck: (a) Truck in Slow Lane; (b) Truck in Fast Lane

Static Axial Stress in Member 02 from 5-Axle Truck: (a) Truck in Slow Lane; (b) Truck in Fast Lane

Maximum Extreme Fiber Stress Range Versus Axial Stress Range for Member 01 at Floortruss 4




xvii

~

62

63

64

66

67

68

69

75

75

80

80

81

81

Figure 52.

Figure 53.

Figure 54.

Figure 55.

Figure 56.

Figure 57.

Figure 58.

Figure 59.

Figure 60.

Figure 61.

Figure 62.

Figure 63.

Figure 64.

Figure 65.





Maximum Extreme Fiber Stress Range Versus Axial Stress Range for Member 04 at Floortruss

Maximum Extreme Fiber Stress Range Versus Axial Stress Range for Member 04 at Floortruss 8·

Maximum Extreme Fiber Stress Rang Versus Axial Stress Range for Member 05 at Floortruss 4








Maximum Extreme Fiber Stress Range Versus Axial Stress Range due to Test Trucks and Random Truck Traffic for Member 01 at Floortruss 4


xviii

82

82

83

83

84

84

85

85

86

86

87

87

90

90

Figure 66.

Figure 67.

Figure 68.

Figure 69.

Figure 70.

Figure 71.

Figure 72.

Figure 73.

Figure 74.

Figure 75.

Figure 76.

Figure 77.

Figure 78.

Figure 79.

















xix

91

91

92

92

93

93

94

94

95

95

96

96

97

97

Figure 80.

Figure 81.

Figure 82.



Critical Extreme Fiber Locations at Floortrusses 4 and 8 (Looking South) 108

Possible Fatigue Crack Locations 113

Assumed Notch Geometry at End of Weld 114

xx

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.


LIST OF TAB LES

Traffic History of the Mobile Delta Crossing Bridges

Floortruss Member Sizes

Filler Plate Weld Cracks at Test Locations

Gages Installed on Diagonal Members

Gages Installed on Bottom Chord Members

Calibration Truck Weights for Quarter-Span Test Location

Calibration Truck Weights for Mid-8pan Test Location

Typical Event Sequence for Calibration Tests

Static Axial Stresses for Test Trucks Parked Directly Over the Gaged Floortruss

Axial and Maximum Extreme Fiber Stress Ranges due to 3-Axle Test Truck Fast Runs in the Slow Lane

Axial and Maximum Extreme Fiber Stress Ranges due to 3-Axle Test Truck Fast Runs in the Fast Lane

Axial and Maximum Extreme Fiber Stress Ranges due to 5-Axle Test Truck Fast Runs in the Slow Lane

Axial and Maximum Extreme Fiber Stress Ranges due to 5-Axle Test Truck Fast Runs in the Fast Lane

Impact Effect of 5-Axle Test Truck in Slow Lane

Stress Ranges at the Neutral Axis of Members 01 and D2 for 5-Axle Test Truck

Stress Ranges at the Neutral Axis of Members 01 and D2 for 3-Axle Test Truck

Effective Stress Ranges at the Neutral Axis of Members 01 and D2 (Based on 250 Random Trucks)

xxi

PAGE

11

15

19

33

33

40

40

42

60

71

71

72

72

73

77

77

77

Table 18.

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.



Comparison of Measured and Calculated Static Axial Stresses due to 5-Axle Truck in Slow Lane over the Gaged Floortruss

Comparison of Measured and Calculated Static Axial Stresses due to 5-Axle Truck in Fast Lane over the Gaged Floortruss

Comparison of Measured and Calculated Static Axial Stresses due to 3-Axle Truck in Slow Lane over the Gaged Floortruss

Comparison of Measured and Calculated Static Axial Stresses due to 3-Axle Truck in Fast Lane over the Gaged Floortruss

Comparison of Measured and Calculated Dynamic Axial Stresses due to 5-Axle Truck in Slow Lane

Comparison of Measured and Calculated Dynamic Axial Stresses due to 5-Axle Truck in Fast Lane

Comparison of Measured and Calculated Dynamic Axial Stresses due to 3-Axle Truck in Slow Lane

Comparison of Measured and Calculated Dynamic Axial Stresses due to 3-Axle Truck in Fast Lane

Effective Stress Ranges at Extreme Fiber Locations

Maximum Stress Ranges at Extreme Fiber Critical Locations

xxii

100

100

101

101

103

103

104

104

107

109

Volume III: Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges - Floortruss-Girder Connections

TABLE OF CONTENTS

Chapter 1 INTRODUCTION BACKGROUND BRIDGE DESCRIPTION LOCATIONS AND EXTENT OF CRACKING PROJECT OBJECTIVES METHODOLOGY AND SCOPE OF REPORT

1

Chapter 2 LITERATURE REVIEW 11 ASSESSMENT OF RETROFIT TYPES STATISTICAL REVIEW

Chapter 3 SELECTION OF RETROFIT PROCEDURE 31

Chapter 4 TEST LOCATIONS 37

Chapter 5 INSTRUMENTATION AND DATA ACQUISITION 49 WORKING ENVIRONMENT INSTRUMENTATION TRANSDUCER POSITIONS DATA ACQUISITION

Chapter 6 DATA COLLECTION AND REDUCTION 82 CALIBRATION TESTS RANDOM TRUCK TESTS DATA REDUCTION EFFECTIVE STRESS RANGE

Chapter 7 RESULTS AT FLOORTRUSS CONNECTIONS BEFORE RETROFIT 102 BEHAVIOR AT TEST LOCATIONS STRESSES FROM STATIC TESTS . CRITICAL LOCATIONS AND CALCULATION OF WEB GAP STRESSES EFFECTIVE STRESS RANGES FROM TRUCK TRAFFIC OUT-OF-PLANE DISPLACEMENTS POTENTIAL FOR CRACKING

Chapter 8 RESULTS AT CONNECTIONS AFTER RETROFIT 174 BEHAVIOR AT TEST LOCATION WEB GAP STRESSES FROM STATIC TESTS EFFECTIVE STRESS RANGES FROM TRUCK TRAFFIC OUT-O~-PLANE DISPLACEMENTS EFFECTIVENESS OF RETROFITS

xxiii


TABLE OF CONTENTS


REFERENCES

xxiv

229

234

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

Figure 15.


LIST OF FIGURES

Typical Floortruss-Girder Connection Not at Pier

Illustration of Out-of-Plane Distortion: (a) In the Web Gap Region; (b) Due to Deflections of the Transverse Floortruss

Typical Fatigue Cracks Found at Floortruss-Girder Connections

Examples of Friction Type Retrofits: (a) Wedges; (b) Compressed Plates

Exaniples of Rigid Attachment Type Retrofits: (a) Welded; (b) Through Bolted

Examples of Rigid Attachment Type Retrofits: (a) Studded; (b) Bracketed Through a Hole in the Girder Web

Examples of Loosening Type Retrofits: (a) Slotted Web; (b) Lengthened Web Gap

Basic Welded Retrofit Detail Chosen for Investigation by Koob et al. (1985)

Basic Studded Retrofit Detail Chosen for Investigation by Koob et al. (1985)

Basic Lengthened Web Gap Retrofit Detail Chosen for Investigation by Koob et al. (1985)

Idealized Histograms Illustrating Common Shapes of Frequency Distributions: (a) Uniform; (b) U-Shaped; (c) Reverse J-Shaped; (d) Bell-Shaped; (e) Skewed Right

Relative Values of Mode, Median, and Mean for Differing Shapes of Frequency Distributions: (a) Symmetrical; (b) Skewed Right; (c) Skewed Left

Shape and Characteristics of Rayleigh Probability Density Curves

Varying Degrees of Correlation: (a) Zero Correlation; (b) Low Correlation; (c) Moderate Correlation; (d) High Correlation

Typical Retrofit Procedure for Connection Plate Slot Retrofit

xxv

EOOE

4

6

8

13

15

16

19

20

21

22

24

27

28

30

33

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Figure 29.



Ten Inch Slot in the Connection Plate and Position of the Three Bolts Removed for the Second Trial Retrofit

Six Inch Slot in the Connection Plate and Position of the Two Bolts Removed for the Second Trial Retrofit

Floortruss-Girder Test Locations on Mobile Delta Crossing Bridges

Position of Test Piers Along the Respective Continuous Spans

Plan View Showing Connections Tested Before and After Retrofit Along the Respective Single Spans of Each Continuous Span

Cross-Section of Bridges at Floortruss-G.irder Connections Away From Piers (Looking North)

Typical Web Gap Region at Floortruss-Girder Connections Away From Piers: (a) Elevation of Inside Face of Girder; (b) Section View Looking in the Longitudinal Direction of the Bridge

Plan View. (with Concrete Deck Removed) of Work Platforms and Scaffolds Used at Pier S3-11 Test Locations

Plan View (with Concrete Deck Removed) of Work Platforms and Scaffolds Used at Pier S3-11 Test Locations

Typical LVDT Arrangement at Floortruss-Girder Connection

Numbering and Locations of Strain Gages on Girder Web Before Retrofit

Numbering and Locations of Strain Gages on Girder Web After Retrofit at the Outside Girder Connection of the First Floortruss South of Pier S3-11

Numbering and Locations of Strain Gages on Girder Web After Retrofit at the Outside Girder Connection of the Second Floortruss North of Pier S3-4

Numbe~ing and Locations of Strain Gages on the Connection Plate Before Retrofit

xxvi

34

35

38

39

41

43

44

50

51

57

59

61

62

63

Figure 30.

Figure 31.

Figure 32.

Figure 33.

Figure 34.

Figure 35.

Figure 36.

Figure 37.

Figure 38.

Figure 39.

Figure 40.

Figure 41.

Figure 42.

Figure 43.



Numbering and Locations of Strain Gages on the Connection Plate at the Outside Girder Connection of the First Floortruss South of Pier S3-11 After Retrofit 65

Numbering and Locations of Strain Gages on the Connection Plate at the Outside Girder Connection of the Second Floortruss North of Pier S3-4 After Retrofit 66

Typical Gage and Fill Plate Locations on Floortruss Members at Pier S3-11 Locations (Looking North) 68

Typical Gage and Fill Plate Locations on Floortruss Members at Pier S3-4 Locations (Looking North) 70

Gage Locations on the Neutral Axes of the Truss Members: (a) Top Chord; (b) End Diagonal; (c) Bottom Chord 71

Numbering and Locations of In-Plane Bending Gages on Girder Flanges 72

Environmental Chamber 80

Trucks Used for Calibration Tests: (a) 3-Axle Truck (Dump Truck); (b) 5-Axle Truck (Lowboy) 85

Test Lanes A, B, and C Used for Calibration Tests (Looking South at Southbound Lanes) 85

Span Positions Used for Static Calibration Tests at Pier S3-11 Locations 91

Span Positions Used for Static Calibration Tests at Pier S3-4 Locations 91

Truck Axle Positioning for Static Calibration Tests: (a) 3-Axle Truck (Dump Truck); (b) 5-Axle Truck (Lowboy) 92

Illustration of Peak-to~Peak Method 97

Typical Distribution of Horizontal Stress in the Connection Plate (at the Outside Girder of the First Floortruss South of Pier S3-11 Before Retrofit) 107

xxvii

Figure 44.

Figure 45.

Figure 46.

Figure 47.

Figure 48.

Figure 49.

Figure 50.

Figure 51.

Figure 52.

Figure 53.

Figure 54.

Figure 55.



Interaction of Connection Plate and Girder Web Plate: (a) Idealized Distribution of Horizontal Stress in Connection Plate; (b) Distortion in Web Gap Region 109

Static Stress Distribution on Outside Girder at the First Floortruss South of Pier S3-11 (106 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 111

Static Stress Distribution on Inside Girder at the First Floortruss South of Pier S3-11 (106 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 111

Static Stress Distribution on Outside Girder at the Second Floortruss South of Pier S3-11 (106 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 112

Static Stress Distribution on Outside Girder at the First Floortruss North of Pier S3-4 (126 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 112 .

Static Stress Distribution on Outside Girder at the Second Floortruss North of Pier S3-4 (126 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 113

Static Stress Distribution on Inside Girder at the Second Floortruss North of Pier S3-4 (126 in. Girder) Before Retrofit with 3-Axle Truck in Outside Lane 113

Static Axial Stress in End Diagonals of the First Floortruss South of Pier S3-11 Before Retrofit with 3-Axle Truck in Lane A or C 115

Static Axial Stress in End Diagonals of the Second Floortruss North of Pier S3-4 Before Retrofit with 3-Axle Truck in Lane A or C 115

Static Stress at the Outside Web-Flange Weld at the First Floortruss South of Pier S3-11 Before Retrofit with 3-Axle Truck in Lane A or C 116

Static Stress at the Outside Web-Flange Weld at the Second Floortruss North of Pier S3-4 Before Retrofit with 3-Axle Truck in Lane A or C 116

Static Stress at the Top of the Connection Plate Weld at the First Floortruss South of Pier S3-11 Before Retrofit with 3-Axle Truck in Lane A or C 117

xxviii

Figure 56.

Figure 57.

Figure 58.

Figure 59.

Figure 60.

Figure 61. (

Figure 62.

Figure 63.

Figure 64.

Figure 65.

Figure 66.

Figure 67.



Static Stress at the Top of the Connection Plate Weld at the Second Floortruss North of Pier S3-4 Before Retrofit with 3-Axle Truck in Lane A or C 117

Gage ON1 Strain Record that Contributed to the Maximum Extrapolated Stress Range Measured for Random Trucks 126

Gage ON2 Strain Record that Contributed to the Maximum Extrapolated Stress Range Measured for Random Trucks 126

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Outside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 129

Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Outside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 129

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Inside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 130

Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Inside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 130

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Outside Girder at the Second Floortruss South of Pier S3-11 Before Retrofit 131

Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Outside Girder at the Second Floortruss South of Pier S3-11 Before Retrofit 131

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Outside Girder at the First Floortruss North of Pier S3-4 Before Retrofit 132

Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Outside Girder at the First Floortruss North of Pier S3-4 Before Retrofit 132

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Outside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 133

xxix

Figure 68.

Figure 69.

Figure 70.

Figure 71.

Figure 72.

Figure 73.

Figure 74.

Figure 75.

Figure 76.

Figure 77.

Figure 78.

Figure 79.

Figure 80.



Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Outside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 133

Web Plate Stress Ranges at the Top of the Connection Plate Weld on the Inside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 134

Web Plate Stress Ranges at the Outside Web-Flange Weld Toe on the Inside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 134

Out-of-Plane Displacement Ranges at the Outside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 146

Out-of-Plane Displacement Ranges at the Inside Girder at the First Floortruss South of Pier S3-11 Before Retrofit 146

Out-of-Plane Displacement Ranges at the Outside Girder at the Second Floortruss South of Pier S3-11 Before Retrofit 147

Out-of-Plane Displacement Ranges at the Outside Girder at the First Floortruss North of Pier S3-4 Before Retrofit 147

Out-of-Plane Displacement Ranges at the Outside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 148

Out-of-Plane Displacement Ranges at the Inside Girder at the Second Floortruss North of Pier S3-4 Before Retrofit 148

Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-of-Plane Displacement Range at the Outside Girder of the First Floortruss South of Pier S3-11 Before Retrofit 149

Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-of-Plane Displacement Range at the Outside Girder of the First Floortruss South of Pier S3-11 Before Retrofit 149

Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-of-Plane Displacement Range at the Inside Girder of the First Floortruss South of Pier S3-11 Before Retrofit 150

Web PI~te Stress Range at the Outside Web-Flange Weld Toe Versus Out-of-Plane Displacement Range at the Inside Girder of the First Floortruss South of Pier S3-11 Before Retrofit 150

xxx

Figure 81.

Figure 82.

Figure 83.

Figure 84.

Figure 85.

Figure 86.

. Figure 87.

Figure 88.

Figure 89.

Figure 90.

Figure 91.



Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-ot-Plane Displacement Range at the Outside Girder of the Second Floortruss South of Pier S3-11 Before Retrofit

Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-ot-Plane Displacement Range at the Outside Girder of the Second Floortruss South of Pier S3-11 Before Retrofit

Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-of-Plane Displacement Range at the Outside Girder of the First Floortruss North of Pier S3-4 Before Retrofit

Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-ot-Plane Displacement Range at the Outside Girder of the First Floortruss North of Pier S3-4 Before Retrofit

Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-of-Plane Displacement Range at the Outside Girder ot the Second Floortruss North of Pier S3-4 Before Retrofit

Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-ot-Plane Displacement Range at the Outside Girder of the Second Floortruss North of Pier S3-4 Before Retrofit

Web Plate Stress Range at the Top of the Connection Plate Weld Versus Out-ot-Plane Displacement Range at the Inside Girder of the Second Floortruss North ot Pier S3-4 Betore Retrofit

Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-of-Plane Displacement Range at the Inside Girder of the Second Floortruss North ot Pier S3-4 Before Retrofit

Stress Ranges at the NS 1 Gage on the Connection Plate at the Outside Girder of the Second Floortruss North of Pier S3-4 Before Retrofit

Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Outside Girder of the First Floortruss South of Pier S3-11 Before Retrofit

Vertical ,Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Inside Girder of the First Floortruss South of Pier S3-11 Betore Retrotit

xxxi

E8Gf.

151

151

152

152

153

153

154

154

159

161

161



Figure 92. Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Outside Girder of the Second Floortruss South of Pier S3-11 Before Retrofit

Figure 93. Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Outside Girder of the First Floortruss North of Pier S3-4 Before Retrofit

Figure 94. Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Outside Girder of the Second Floortruss North of Pier S3-4 Before Retrofit

Figure 95. Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the Inside Girder of the Second Floortruss North of Pier S3-4 Before Retrofit

Figure 96. Effective Stress Ranges at Floortruss-Girder Connections Transposed on Laboratory Results From Fatigue Tests of Transverse Connections by Fisher et al. (1990)

Figure 97. Typical Distribution of Horizontal Stress in the 10 Inch Slotted Connection Plate

Figure 98. Typical Distribution of Horizontal Stress in the 10 Inch Slotted Connection Plate with the Top Three Bolts Removed

Figure 99. Typical Distribution of Horizontal Stress in the 6 Inch Slotted Connection Plate

Figure 100. Typical Distribution of Horizontal Stress in the 6 Inch Slotted Connection Plate with the Top Two Bolts Removed

Figure 101. Static Stress Distribution at the 10 Inch Slotted Connection with 3-Axle Test Truck in Outside Lane

Figure 102. Static Stress Distribution at the 6 Inch Slotted Connection with 3-Axle Test Truck in Outside Lane

Figure 103. Web Plate Stress Ranges at the Outside Web-Flange Weld Toe at the 10 Inch Slotted Connection Before Retrofit

xxxii

162

162

163

163

171

177

177

178

178

184

185

198



Figure 104. Web Plate Stress Ranges at the Outside Web-Flange Weld Toe at the 10 Inch Slotted Connection After Retrofit

Figure 105. Web Plate Stress Ranges at the Outside Web-Flange Weld Toe at the 6 Inch Slotted Connection Before Retrofit

Figure 106. Web Plate Stress Ranges at the Outside Web-Flange Weld Toe at the 6 Inch Slotted Connection After Retrofit

Figure 107. Web Plate Stress Ranges at the Top of the 10 Inch Slotted Connection Plate to Web Weld

Figure 108. Web Plate Stress Ranges at the Top of the 6 Inch Slotted Connection Plate to Web Weld

Figure 109. Web Plate Stress Ranges at the Side of the 10 Inch Slotted Connection Plate to Web Weld

Figure 110. Web Plate Stress Ranges at the Side of the 6 Inch Slotted Connection Plate to Web Weld

Figure 111. Out-of-Plane Displacement Ranges at the 10 Inch Slotted Connection Before Retrofit

Figure 112. Out-of-Plane Displacement Ranges at the 10 Inch Slotted Connection After Retrofit

Figure 113. Out-of-Plane Displacement Ranges at the 6 Inch Slotted Connection Before Retrofit

Figure 114. Out-of-Plane Displacement Ranges at the 6 Inch Slotted Connection After Retrofit

Figure 115. Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-of-Plane Displacement Range at the 10 Inch Slotted Connection After Retrofit

Figure 116. Web Plate Stress Range at the Outside Web-Flange Weld Toe Versus Out-of-Plane Displacement Range at the 6 Inch Slotted Connection After Retrofit

xxxiii

198

199

199

200

200

201

201

211

211

213

213

214

214



Figure 117. Web Plate Stress Range at the Top of the 10 Inch Slotted Connection Plate to Web Weld Versus Out-of-Plane Displacement Range After Retrofit 217

Figure 118. Web Plate Stress Range at the Top ot the 6 Inch Slotted Connection Plate to Web Weld Versus Out-ot-Plane displacement Range After Retrofit 215

Figure 119. Web Plate Stress Range at the Side of the 10 Inch Slotted Connection Plate to Web Weld Versus Out-of-Plane Displacement Range After Retrofit 216

Figure 120. Web Plate Stress Range at the Side of the 6 Inch Slotted Connection Plate to Web Weld Versus Out-of-Plane Displacement Range After Retrofit 216

Figure 121. Vertical Force Component Range in the End Diagonal Versus Out-ot-Plane Displacement Range at the 10 Inch Slotted Connection After Retrofit 220

Figure 122. Vertical Force Component Range in the End Diagonal Versus Out-of-Plane Displacement Range at the 6 Inch Slotted Connection After Retrofit 220

xxxiv

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.


LIST OF TABLES

Traffic History of the Mobile Delta Crossing Bridges

Web Gap Geometries and Crack Lengths

Plate Girder Dimensions

Floortruss Member Double-Angle Dimensions

Gages Installed at Pier S3-11 Locations Before Retrofit

Gages Installed at Pier S3-4 Locations Before Retrofit

Gages Installed at the Outside Girder Connection of the First Floortruss South of Pier S3-11 After Retrofit

Gage Installed at the Outside Girder Connection of the Second Floortruss North of Pier S3-4 After Retrofit

Calibration Truck Weights for S3-11 Before Retrofit

Calibration Truck Weights for S3-4 Before Retrofit

Calibration Truck Weights for S3-11 After Retrofit

Calibration Truck Weights for S3-4 After Retrofit

Typical Event Sequence for Calibration Tests

Effective Stress Ranges at Critical Locations for Random Truck Data Before Retrofit

Fatigue Limit Exceedance Rates at Critical Locations for Random Truck Data Before Retrofit

Stress and Displacement Ranges at Critical Locations for 5-Axle Test Truck Fast Runs (Typical Loading) Before Retrofit

Stress and Displacement Ranges at Critical Locations for 3-Axle Test Truck Fast Runs Before Retrofit

Displac'ement Range Statistics at Test Locations for Random Truck Data Before Retrofit

xxxv

2

45

47

48

74

75

76

77

86

86

88

88

89

124

137

140

141

144

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.

Table 29.

Table 30.



Regression Analyses of Critical Location Stress Ranges Versus Displacement Ranges for Random Truck Data Before Retrofit

Regression Analyses of Vertical Force Ranges in End Diagonals Versus Displacement Ranges for Random Truck Data Before Retrofit

Static Stresses and Displacements at the 10 Inch Slotted Connection, Before and After Retrofits with Truck Directly Above Floortruss

Static Stresses and Displacements at the 6 Inch Slotted Connection, Before and After Retrofits with Truck Directly Above Floortruss

Percentage Change a of Static Stresses and Displacements at Critical Locations After Retrofit at the 10 Inch Slotted Connection with Trucks Directly Above Floortruss

Percentage Change a of Static Stresses and Displacements at Critical Locations After Retrofit at the 6 Inch Slotted Connection with Trucks Directly Above Floortruss

Effective Stress Ranges at Critical Locations for Random Truck Data Before and After Retrofit

Percentage Change a of Effective Stress Ranges at Critical Locations After Retrofit for Random Truck Data at the 10 and 6 Inch Slot Locations

Stress and Displacement Ranges at Critical Locations at the 10 Inch Slotted Connection for Test Truck Fast Runs, Before and After Retrofit

Stress and Displacement Ranges at Critical Locations at the 6 Inch Slotted Connection for Test Truck Fast Runs, Before and After Retrofit

Percentage Change a of Stress and Displacement Ranges at Critical Locations After Retrofit at the 10 Inch Slotted Connection for Test Truck. Fast Runs

Percentage Change a of Stress and Displacement Ranges at Critical Locations After Retrofit at the 6 Inch Slotted Connection for Test Truck Fast Runs

xxxvi

157

164

181

182

190

191

194

195

203

204

205

206

Table 31.

Table 32.

Table 33.

Table 34.



Stress and Displacement Ranges for Estimated Side-by-Side Test Truck Fast Runs at Critical Locations at the 10 and 6 Inch Slotted Connections with Fully Bolted Connection Plates

Displacement Range Statistics at Test Locations Before and After Retrofit for Random Truck Data

Regression Analyses of Critical Location Stress Ranges Versus Displacement Ranges for Random Truck Data Before and After Retrofit

Regression Analyses of Vertical Force Ranges in End Diagonals Versus Displacement Ranges for Random Truck Data Before and After Retrofit

xxxvii

208

210

217

221

Volume IV: Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges - Executive Summary

TABLE OF CONTENTS

INTRODUCTION

FLOORBEAM-GIRDER CONNECTIONS Rigid Attachment Repair Summary and Conclusions Recommendations

FLOORTRUSS-GIRDER CONNECTIONS Selection of a Retrofit Procedure Summary and Conclusions Recommendations

FILLER PLATE WELD CRACKS Evaluation Methodology Summary and Conclusions Recommendations

REFERENCES

xxxviii

2

9

~5

19

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Volume IV: Evaluation of Fatigue Cracking in 1-65 Mobile Delta Crossing Bridges - Executive Summary

LIST OF FIGURES

Floorbeam-Girder Connection: (a) Section through Plate Girder at Support; (b) Out-ot-Plane Distortion ot Web Gap

Typical Fatigue Cracks Found at Floorbeam-Girder Connections

Rigid Attachment ot Connection Plate to Girder Flange: (a) Section through Plate Girder; (b) Section through Floorbeam Knee Brace

Section through Plate Girder at Floortruss-Girder Connection

Typical Fatigue Cracks Found. at Floortruss-Girder Connections

Connection Plate Slot Detail

Typical Filler Plate Weld Crack

xxxix

3

5

6

10

10

12

16

INTRODUCTION

The 1-65 Mobile Delta Crossing Bridges constitute a six mile section (northbound and

southbound) of interstate highway over the Mobile River delta plains located approximately 20

miles north of Mobile, Alabama. The bridges consist of three basic types: precast concrete

spans, steel plate girder spans, and twin steel tied arch spans. The steel plate girder spans

are made of two main girders which span in the direction of the roadway with transverse

members spaced at 25 ft between the girders. The transverse member types vary based on

the depth of the main girders and are either built-up floorbeams or floortrusses. Three

stringers span continuously across the top of the transverse floorbeams or floortrusses. The

top flange of the girders and stringers support the composite concrete deck. At the tied

arches, the stringers and deck are supported by transverse floortrusses. These floortrusses

are spaced at 50 ft and span between the tie girders of an arch on each side of the roadway.

All steel in the plate girder spans and the tied arch spans is ASTM A588 Grade 50W

weathering steel.

The bridges were constructed in the early 1980's and were opened to traffic in 1983.

The traffic history indicates that approximately 5 million heavy commercial vehicles (3 or more

axles) had crossed each bridge span by the end of 1993. In 1990 and 1991, Alabama

Highway Department (AHD) bridge inspectors discovered fatigue cracks in the girder webs at

floorbeam-girder connections at six locations. All these locations were at girder supports at

piers along the steel plate girder spans. Fatigue cracks were also discovered in welds at five

floortruss-girder connections not at piers. Because the plate girder spans are a two-girder

fracture critical system, there was considerable motivation for providing effective repairs to the

fatigue cracks. The connection details where the fatigue problems were identified are also

repeated a large number of times along the bridges. There are 192 floorbeam-girder

connections at piers like the ones where fatigue cracking was found and 768 additional

similar connections not at piers. There are also 144 floortruss-girder connections not at piers

of the type where the fatigue cracking of the welds was found. Because of the repetition of the

details and the possibility of wide spread fatigue problems in the future, efficient repair

methods were desirable.

1

A connection repair was designed by the AHD for use at the floorbeam-girder

connections. The repair was originally considered for possible use as a preventative

maintenance measure at all 192 floorbeam-girder connections at piers. However, because of

the large expense of performing the repair and because the effectiveness of the repair could

not be guaranteed, field testing of the repair was necessary. The goal of the field tests, which

were a part of this research project, was to evaluate the stress conditions which caused the

fatigue cracking and to evaluate the effectiveness of the AHD repair method.

The fatigue cracks at the floortruss-girder connections were short cracks through the

ends of the fillet welds between the stiffener-connection plate and the girder web. An obvious

short term solution to this problem was to simply grind out the cracked welds. However, the

cracking was suspected to be an early indication of more serious girder web cracking that

might occur in the future. One objective of the project reported here was to evaluate the stress

conditions that caused the cracks and recommend an appropriate repair method.

A third problem was also identified at the floortrusses of the tied arch spans. A number

of welds at spacer plates in the double angle floortruss members were found to be cracked.

These cracks were parallel to the applied stress on the members and did not pose an

immediate threat. However, the possibility of future cracking in the tension members was a

concern. One maintenance strategy considered by the AHD was to remove the filler plates

and grind out the welds. However, this option would be very expensive and was possibly

unnecessary. For the tension members alone this would require the removal of approximately

960 filler plates. A prime objective of the project reported here was to measure the traffic

loading stresses in the floortruss tension members and to address whether it is necessary to

remove the filler plates from these members and grind out the welds:

FLOORBEAM-GIRDER CONNECTIONS

The steel plate girder bridges with transverse floorbeams span across the Tensaw

River, Miffin Lake, Middle River, and Little Lizard Creek. These are three and four span

continuous sections with plate girder web depths of either 84 in. or 96 in. The floorbeam

girder connections at the girder supports are made of a vertical connection plate welded to the

girder web and bolted to the floorbeam (see Figure 1). The connection plate on the inside and

2

DETAILA

I--__ -~-.L-I .... :....J-- Plate Girder Web

Floorbeam -~

Connection Plate

(a)

Flange

Conn. PI.

DETAIL A

(b)

Transverse Stiffener

Girder Support

Pier

Web Gap

Figure '1. Floorbeam-Girder Connection: (a) Section through Plate Girder at Support; (b) Out-of-Plane Distortion of Web Gap

3

stiffener on the outside face of the web are welded to the girder web and compression

(bottom) flange and tight fit against the tension (top) flange. The weld between the connection

plate and girder web terminates a short distance away from the tension flange of the girder.

The area between the end of the weld and the tension flange is referred to as the web gap.

Because the connection plate is not welded to the tension flange, the web gap is

subjected to out-of-plane distortion of the girder web resulting from end rotation of the

floorbeam (see Figure 1). As traffic crosses the floorbeam, the end of the floorbeam rotates

causing the top of the connection plate to be pulled inward and producing the out-of-plane

distortion of the web which results in distortion-induced fatigue stresses in the web. An

efevation of the inside of the girder web at the top of the floorbeam connection plate is shown

in Figure 2 to illustrate the locations of typical cracks. Cracking was found at a total of six

connections. Five of the connections with cracks were under the southbound lanes. Three of

these connections were at the girder on the slow lane side of the bridge, while the other two

occurred at the girder on the fast lane side.

Rigid Attachment Repair

The AHD developed the retrofit detail shown in Figure 3 to arrest crack growth at the

six connections with fatigue cracks. The retrofit detail is referred to as "rigid attachment"

because the floorbeam connection plate is rigidly attached to the tension flange of the girder.

The purpose of the retrofit is to reduce out-of-plane displacement at the top of the connection

plate so that the associated distortion-induced stresses are no longer large enough to cause

fatigue cracks to form or propagate.

Summary and Conclusions

The potential for future fatigue cracking at additional floorbeam-girder connections and

the effectiveness of the rigid attachment repair were evaluated through field testing. Field

measurements were made to determine the magnitudes of the out-of-plane displacements at

the top of the connection plate and distortion-induced stresses in the girder web at typical

floorbeam-girder connections. Measurements were made with loading from trucks of known

weight and with loadin~ from normal truck traffic crossing the bridges. Field measurements of

stresses and displacements were made at a total of ten connections, five at each of the two

4

fatigue crack at Jf web-flange weld

girder tension flange

weld

~--~------~~---

/ fatigue cracks at connection plate weld

~ Q. c: o :a Q) c: c: o (.)

girder web

Figure 2. Typical Fatigue Cracks Found at Floorbeam-Girder Connections

5

Reinf. Not Shown FA ::(·~:>::~-:~E----~<:) ~".: ~ ..... :: .

1/2" to 1" Deep Sawcut (Typ.)

©©: I

©©: I

.©©: I

©©:

(a)

2' - 6" ± (Typ.)

Reinf. Not Shown

. . .. .. .... . .

• ~', e' " • : ••

L 8x8x5/8x6-3/4" or 5/8" Bent If.

. '.'

SECTION A-A

(b)

Figure -3. Rigid Attachment of Connection Plate to Girder Flange: (a) Section through Plate Girder; (b) Section through Floorbeam Knee Brace

6

girder depths. These five connections included three connections at girder supports at piers,

one connection at the first floorbeam away from the pier, and one connection at the second

floorbeam away from the pier. The field measurements were compared to results of laboratory

tests performed by other researchers to assess the potential for future fatigue cracking at

similar connections along the bridge. Measurements were also made at four of these

floorbeam-girder connections after the rigid attachment retrofit was performed. These

measurements provided direct comparisons for evaluating the effectiveness of the rigid

attachment retrofit.

The field test data indicated that only two of the six connections tested at pier

floorbeam locations experienced distortion-induced stress ranges large enough to cause

fatigue cracking. One of the connections was at an 84 in. girder, and one was at a 96 in.

girder. Given the fact that cracks have been detected at the piers at four locations along 84 in.

girders and one location along 96 in. girders, and that low stress conditions were found at four

of the connections tested, it appears that significantly different stress conditions exist at the

pier floorbeam connections along the bridges. No pattern was identified that clearly indicates

which locations along the bridges are most likely to experience cracking in the future. The test

results do indicate that cracking is not likely to occur at all pier floorbeam locations. The small

number of connections chosen for field tests does not allow an accurate estimate of

percentage of connections at which cracking will occur. However, it appears that cracking can

be expected at approximately one-third of the pier f100rbeam connections.

Both of the connections tested at the first floorbeam away from pier supports had high

distortion-induced stresses. The stress ranges occurring at both connections were high

enough to eventually caus~ fatigue cracking. Cracking can be expected in the future at

connections at the first floorbeam away from the pier.

The connection details tested at the second floorbeam away from the pier had very

small distortion-induced stresses. Fatigue cracking is very unlikely to occur at any such

locations in the future.

Since the plat~ girders of a two girder system are considered fracture critical members,

the possibility of the web cracks turning and propagating into the tension flange is of interest.

7

The field measurements indicate that for all the details tested, the in-plane stress ranges in the

tension flange were very small (below 1.0 ksi) at the piers and first floorbeams away from the

piers. Therefore, the initiation of a web gap crack and growth to a critical length into the

tension flange is not likely to occur in a short period of time. Hence, bridge inspections are not

necessary at times shorter than the normal two year inspection interval.

An evaluation of the effectiveness of the rigid attachment retrofit was performed by

comparing distortion-induced stress ranges and displacements in uncracked connections

before and after the retrofit was performed. The comparisons indicated that the positive

attachment retrofit is an effective repair method. The retrofit was installed at one pier location

where previous measurements indicated that fatigue cracking would eventually occur. The

retrofit reduced the out-of-place displacement of the connection plate and distortion-induced

stresses by approximately 70%. This was a large enough reduction in distortion-induced

stresses to conclude that fatigue cracking will not occur at that location in the future. Based on

the field test results, the positive attachment retrofit was judged to be an effective measure to

prevent cracking and an effective repair method for connections where distortion-induced

fatigue cracks have formed.

Recommendations

Due to the low in-plane stress ranges measured in the tension flanges at all the test

locations, web-gap cracking resulting from out-of-plane bending at the floorbeam-girder

connections does not pose as great an emergency as once suspected. Furthermore, the

distortion-induced stress conditions at the various test locations were shown to vary

significantly. These findings indicate that performing the positive attachment retrofit to prevent

future cracking is unwarranted. Instead, the results of this project indicate that floorbeam

girder connections with web gap cracking should be repaired on an as needed basis. The

normal two year inspection cycle will be sufficient for identifying new occurrences of web gap

cracks. When cracks are found as a result of inspection, the crack tips should be removed by

drilling a one inch diameter hole at each crack to temporarily arrest the growth of the cracks.

This period of arrest will allow all the connections with web gap cracking found in two

8

consecutive inspections to be repaired by rigid attachment at the same time. However, it is not

recommended that hole drilling be used to delay rigid attachment for more than three years.

FLOORTRUSS-GIRDER CONNECTIONS

The deepest plate girder spans along the Mobile Delta Crossing Bridges have

transverse floortrusses supporting the stringers and deck instead of built up floorbeams.

Floortrusses were used at girder web depths of 106 in., 126 in. and 148 in. The floortruss

girder connection is composed of a vertical stiffener-connection plate that is welded to the

plate girder web and bottom flange and bolted to the floortruss through a trapezoidal gusset

plate (see Figure 4). The floortruss connection plates are welded to the girder web and the

compression flange (bottom flange) and are tight fit to the tension flange (top flange) in the

negative moment regions of the main girders near the pier supports. The connection plate to

girder web welds terminate a short distance from the girder tension flange plate which creates

a web gap similar to the one described at the floorbeam-girder connection. This web gap

region also experiences distortion-induced fatigue stresses which result from out-of-plane

displacement at the top of the connection plate.

Fatigue cracks were found by AHD bridge inspectors at floortruss-girder connection

web gaps through the top of the connection plate to web welds as shown in Figure 5. The

cracks had not propagated from the weld throats into the web plate, and therefore, did not

pose an immediate threat to the bridge integrity. However, this type of crack is likely to

eventually propagate into the girder web. The crack could then turn perpendicular to the in

plane bending stress in the girder web and become a much more significant problem.

Cracking was discovered at a total of five floortruss-girder connections. Unlike the cracking of

the floorbeam-girder connections, all cracks were located at connections not at girder

supports. All cracks found were in the negative moment region of the girders in the web gap

near the tension flange at connections one and two floortrusses away from the girder

supports. Three of the five connections with fatigue cracking were chosen for investigation

through field measurements. In addition, three connections with no fatigue cracks were also

chosen for investigation.

9

connection plate ~ moves inward

' .. :- ._-, ,-- ,-.

: ~.

Figure 4. Section through Plate Girder at Floortruss-Girder Connection

girder web

girder tension flange

weld

fatigue cracks in connection

plate weld

Figure 5. Typical Fatigue Cracks Found at Floortruss-Girder Connections

10

Selection of a Retrofit Procedure

The fatigue cracks could easily be eliminated by grinding out the ends of the welds.

However, data taken early on during the field tests revealed that stress ranges in the girder

web plates at these floortruss connections far exceeded the fatigue limits for distortion

induced cracking of the web. Hence, grinding out the cracks at the ends of the welds would

solve the immediate fatigue problem, but would not address the long-term problem of cracking

in the girder webs.

A positive attachment retrofit was the preferred retrofit at floorbeam connections

discussed earlier. The retrofit was found to be effective in reducing the distortion-induced

stresses to acceptable levels. It appears reasonable that if the rigid attachment retrofit was

used at the floortruss-girder connections, it would be effective there as well. However, this

retrofit method is very expensive to implement. The expense prompted the AHD to consider

other possible retrofit procedures. After reviewing preliminary results from the field research

obtained in July 1992, the AHD, together with Auburn University, developed two trial retrofit

details to be implemented at two locations.

The trial retrofits were designed to make the connection more flexible by lengthening

the web gap which was predicted to reduce the cyclic distortion-induced stresses in the web

gap region. Fisher et al. (1990) investigated the concept of lengthening the web gap by

removal of a full width piece of the connection plate, and it was found to be effective. He

concluded that the available data was insufficient to recommend a widely applicable retrofit

procedure and stated that care should be taken to prevent web gap distortion from increasing

substantially when implementing this method.

For the 1-65 bridges, removal of a full width piece of the top of the connection plate

would provide a condition that would be difficult to repair if the retrofit was unsuccessful.

Rather than removing the entire width of the plate, the web gap was lengthened by creating a

slot in the connection plate as illustrated in Figure 6. Enough plate was left so that a positive

attachment retrofit could still be performed if the slot retrofit proved unsuccessful. Slot lengths

of 10 in. and 6 in. were chosen for implementation and study. In addition, the effect on

distortion-induced stresses from the removal bolts from the top of the connection between the

11

connection plate

flame-cut edge for ___ -+_~ plate removal

flame-cut to edge of connection plate weld

truss top chord

o

o

o

web-flange weld

portion of connection plate and weld ground smooth with web face

(all cut edges to be ground smooth)

connection plate to web weld

web plate

A325 high strength bolts (23) 7/8" dia. @ 3" c.c.

Figure 6. Connection Plate Slot Detail

12

trapezoidal gusset plate and the stiffener/connection plate was investigated. Two bolts were

removed at the connection with the 6 in. slot, and three bolts were removed at the connection

with the 10 in. slot. Field measurements were made at both these connections after the slot

was cut with all connection plate bolts in place and with selected bolts removed.


Evaluating the potential for future fatigue cracking at additional floortruss-girder

connections through field testing was one objective of the research. Stress and displacement

measurements were taken at six floortruss-girder connection details. These connections were

located one and two floortrusses away from the girder supports at two different girder depths.

The field tests verified that the stress ranges occurring in the connection plate to web welds

were high enough at all six connections test~d to eventually cause fatigue cracking. The field

tests also revealed that all six floortruss-girder connections tested experienced distortion

induced stress ranges sufficient to cause fatigue cracking in the girder webs. Based on the

field test results, fatigue cracking in the connection plate to web welds and girder webs can be

expected to eventually occur at a large percentage of floortruss-girder connections along the

plate girder spans.

Some significant variations in the stress conditions were found at the floortruss-girder

connections tested. Overall the 106 in. girder connections consistently experienced more

severe stress conditions than the 126 in. girder connections. Based on this result, it is

estimated that the stress conditions at the 148 in. girders are less severe than at the smaller

girder depths. The distortion-induced stresses in the girder webs were also higher than

expected. The field measurements indicate that heavy side-by-side trucks can cause yielding

of the web in the web gap at the floortruss girder connections of the 106 in. girders. To reduce

the distortion-induced stresses and improve the fatigue performance of the floortruss-girder

connections, the lengthened web gap retrofit was designed, implemented and tested.

The 6 in. slot retrofit was judged to be ineffective. The 6 in. slot produced only modest

changes in the distortion-induced stress conditions. The removal of bolts from the

connections was also found to be ineffective at both slot lengths. The removal of bolts

13

produced both increases and decreases in stresses at various critical locations. In most cases

the changes were less than 6%.

Overall, the 10 in. slot retrofit was judged to be successful. The 10 in. slot retrofit was

effective in reducing the stress ranges at critical locations by approximately 50 to 75% while

limiting the out-of-plane displacements to an acceptable level. The 10 in. slot retrofit will not

eliminate all possibility of fatigue cracking at the floortruss-girder connections, but the

reductions in stress ranges at the slot correspond to an increase in remaining fatigue life by

factors of 27 to 64.

Recom mendations

The field test results indicate that the distortion-induced stresses at the floortruss-girder

connections are very high at the 106 in. girders. The use of a slot retrofit similar to the one

tested in this project is recommended as a preventative maintenance measure at the

floortruss-girder connections along the 106 in. girders. A slot of the type shown in Figure 6

with a total length of 12 in. is recommended for floortruss-girder connections away from the

girder supports in the negative moment regions. This small increase in slot length, from 10 in.

to 12 in., beyond the length used in the field tests will enhance the performance of the retrofit

without significantly increasing the out-of-plane displacement of the connection. The slot will

eliminate the possibility of yielding in the web gaps and will significantly increase the

remaining fatigue life of the connections. Based on the less severe stress conditions

measured at the 126 in. girders, the need for preventative measures at the 126 in. and 148 in.

girders is not as great as at the 106 in. girders.

The stresses in the connection plate weld at the top of the connection plate were found

to be above the fatigue limit. Weld cracks of the type already found by AHD bridge inspectors

can be expected in the future and will probably be common at all girder depths. As originally

mentioned in the project proposal, simply grinding out the weld cracks is a viable repair.

Although the repair will not be permanent, several years may pass before the weld cracks re

initiate. Considering the large distortion-induced stresses occurring at the floortruss-girder

connections, the slotted connection plate retrofit is recommended instead of simply grinding

14

out the weld cracks. This will provide a significant increase in fatigue life without significantly

increasing the repair costs relative to grinding out the weld cracks.

The slotted connection plate retrofit in combination with drilling holes aUhe crack tips is

recommended as a repair for distortion-induced web cracking at the floortruss-girder

connections. For example, this combination is recommended when web cracking occurs at

connections where the slot was not used as a preventative measure. Although there is no

proof that this will provide a permanent repair, this method should prevent re-initiation of

cracks for several years.

FILLER PLATE WELD CRACKS

Bridge inspectors discovered filler plate weld cracks in numerous double angle tension

and compression members of the floortrusses of the tied arch spans. A typical filler plate weld

crack is shown in Figure 7. The ends of the filler plates are attached to the back to back legs

of the double angles with groove welds. The welds are approximately two inches long and

are ground flush with the ends of the filler plates. The double angles and filler plates were

made of unpainted A588 weathering steel and were susceptible to corrosion. The normal

weathering, or corrosion, of the A588 steel caused rust to build up between the double angle

legs and filler plates. Built up corrosion products in restricted areas, called "pack rust", can

generate pressures of up to 10 ksi (Kulicki et al. 1990). By considering an internal pressure of

10 ksi over the area between the double angle and filler plate and making a generous

estimate of the typical weld throat size, the force on the weld at each end of the filler plate was

calculated to be approximately three times the capacity of the weld. Thus, the confined rust is

considered to be the cause of the filler plate weld cracks in the double angle members at the

tied arch spans.

The weld cracks created by the "pack rust" are parallel to the applied stress on the

double angles and have not propagated into the angles and pose no immediate threat to the

integrity of the trusses. However, the possibility of future cracking in the tension members was

a concern prior to this project. The goal of the project reported here was to measure the traffic

loading stresses in the floortruss tension members and to address whether it is necessary to

remove the filler plates from these members and grind out the welds.

15

Figure 7. Typical Filler Plate Weld Crack

16

Evaluation Methodology

Field measurements of stresses in the floortruss tension members were made at

quarter-span and mid-span locations along the arch span with loading from trucks of known

weight and with loading from the normal truck traffic crossing the bridges.

The field measurements were compared to the fatigue limit for the filler plate weld

details. Stresses in the filler plate weld details of the double angle members at the tied arch

span can be classified under Stress Category C (AASHTO 1991). This category applies to the

base metal with longitudinal loading at fillet and groove welded attachments less than or

equal to two inches in length. The allowable fatigue stress range, or fatigue limit, for a

Category C detail with redundant load paths and subjected to over two million cycles is 10 ksi.

For fatigue cracking to occur in the double angle tension members very near the filler plate

welds, the area around the weld details would have to experience stress ranges from traffic

loading greater than 10 ksi. Along with the comparison of the field test data to the AASHTO

fatigue limit, fracture mechanics approaches were applied to draw conclusions about whether

fatigue cracking will pose a problem at the tied arch span. Fracture mechanics methods were

used to predict the fatigue limit and to estimate a tolerable crack size and fatigue life for the

floortruss double angle members at the tied arch span.


Practically all of the field measurements of stresses were made at two floortrusses

along the arch span. Some measurements were made at a third floortruss for comparison

purposes. The difference in stresses measured at the various floortrusses was found to be

minimal. Thus, the measured stresses were concluded to be representative of those occurring

in all the floortrusses.

Based on field test data, the four outer diagonal members were found to be the highest

stressed members in the floortrusses. For these four critical members, the maximum extreme

fiber stress ranges were 15 to 20 percent higher than the axial stress ranges. These

percentage increases in stresses due to bending were found to be unaffected by truck weight

or lane position. Stresses found in the four critical members using a general planar frame

17

computer program were consistently higher yet comparable in most cases to the measured

stresses.

The highest stress range found through field testing was 4.4 ksi. This worst case value

is less than half the design AASHTO fatigue limit of 10 ksi for the filler plate weld details. The

conditions existing after the filler plate welds have cracked (failed) are not considered to be

significantly worse than the as-designed conditions. These comparisons indicate that fatigue

cracking is not likely to occur at the filler plate welds regardless of whether or not the welds fail

due to the pack rust.

Estimations and predictions concerning the double angle tension members based on

fracture mechanics principles also indicate that fatigue cracking is not a threat to the integrity

of the tied arch span floortrusses. Estimates of the fatigue-crack-initiation threshold indicate

that fatigue cracking will not occur. Also, fatigue life calculations indicate that if fatigue cracks

do initiate in the angles at the filler plate welds, these cracks will grow at a sufficiently slow

rate to be found in the normal two year bridge inspections.

Therefore, based on the comparisons of maximum measured stress range to the

fatigue limit and the results from the fracture mechanics calculations, fatigue cracking in the

double angle tension members near the filler plate weld cracks is not problem.

Recommendations

Because of the conclusion that the development of fatigue cracking in the tension

members of the tied arch span floortrusses is not a problem, removing the filler plates from

those members and grinding out the welds is not necessary. However, if weld cracks become

severe enough to cause consecutive filler plates to break completely away from a

compression member, buckling in that member is possible. Hence, for compression members

having two or more consecutive failed filler plate welds, replacement of the filler plates using a

new filler plate and a single high strength bolt is recommended.

18

REFERENCES

American Association of State Highway and Transportation Officials (1992). Standard Specifications fQr Highway Bridges, 15 Edition, Washington, D.C.

Fisher, J.W., Jin, J., Wagner, D.C., and Yen, B.T. (1990). "Distortion Induced Cracking in Steel Bridge Members," Final Report. ATLSS No. 90-07. NCHRP No. 12-28(6}.

Kulicki, J.M., Prucz, Z., Sorgenfrei, D.F., Martz, D.R., and Young, W.T. (1990). "Guidelines for Evaluating Corrosion Effects in Existing Steel Bridges." Transportation Research Record ~, National Research Council, Washington, D.C., 7-19.

19

sponsored - Auburn University

Documents