Investigation of Distortion-Induced Fatigue Cracking in a Seismically-Retrofitted Bridge Riyadh Hindi, PhD, PEng Will Lindquist, PhD, PE Ahmed Ibrahim,

Investigation of Distortion-Induced Fatigue Cracking in a Seismically-Retrofitted Bridge

Riyadh Hindi, PhD, PEng Will Lindquist, PhD, PE

Ahmed Ibrahim, PhD, PEYing Tung, PhD Candidate

Presentation Outline

• Brief Introduction and Background• Linear Elastic Finite Element Analysis

• Moving Live Loads• Stationary Live Loads• Temperature Loads

• Preliminary Fatigue Calculations• Potential Repair Options• Field Monitoring• Summary and Conclusions

Project Location

Section 1

Section 2

Section 3Section 4

Section 3:• AADT approximately 94k, 14k trucks

• Composite reinforced CIP deck

• Cracks found on interior girders in the web- gap region near new diaphragms

Retrofit Plan & Cross-Section Views

Retrofit Plan & Cross-Section Views

Retrofit Plan & Cross-Section Views

Crack Locations

*Cracks are located on interior girders

Finite Element Analysis

• Loads / Load Application• Dead Load• Live Load• Temperature Load

• Finite Element Models (FEMs) evaluated to date• SAP2000• Abaqus

• Future Work

Concrete Deck

Longitudinal Steel Girders

Rigid Links

Shared nodes in Stiffener-Web intersection

Web-gap region

Top flange

Connection plate (stiffener)

Loading

• Dead Load: Total weight of the structure

• Temperature Load : ± 80˚F applied to the entire model

• Moving Load / Point Truck Load : Exterior girder Interior girder

Load Applied

Point Loads / Moving Loads

4k4k

16k16k

16k16k

Bent 64

320 lb/fton both

* Wheels are 6 ft. apart

Point Loads / Moving Loads

FEMs Developed with SAP2000

Loads• Temperature Loads• Moving Loads + Lane Load• Point Load + Lane Load

Models• Original• Retrofit• Retrofit Options

G5G4G3G2G1

SAP2000 Model List

Original Model Retrofit (Fixed braces)

Retrofit (Braces free to

rotate)

Retrofit (Weld web gap at

crack location)

Retrofit(Remove retrofit

links at B64)

Dead load Moving Load (G1) Moving Load (G2) Moving Load (G4) Moving Load (G5)

Temperature Load (±80˚F) Point Truck + Lane Load (G1) Point Truck + Lane Load (G2) Point Truck + Lane Load (G4) Point Truck + Lane Load (G5)

SAP2000 Model

Bent 61

Bent 62

Bent 63

Bent 64

Retrofit Cross-Bracing left of Bent 64

Bent 64 -• Crack Locations near B64 have slightly higher stresses• Fine mesh and shell elements for the bracing at B64.

Temperature Load = 80° F | B64

Average Principal Tensile Stress

Average Principal Tensile Stress

Truck Truck

Near B63 Near B64

Truck Point Load + Lane Load

Truck

Truck Point Load vs. Moving Load | B64

Moving load on G2 Truck + Lane load on G2

Truck

Average Principal Tensile Stress

Interior vs. Exterior Girders | B64

Average Principal Tensile Stress

Truck + Lane load on G1 Truck + Lane load on G2

TruckTruck

Vertical Deflection | Differential Deflection

• Girder Deflections• Differential Girder Deflections• Out-of-Plane Distortion

Girder Deflections

Vertical Deflection | Truck Load

Vertical Differential Deflection

Differential Girder Deformation

Out-of-Plane Displacement

Y-axis

“X-Bracing” “K-Bracing” “X-Bracing”

Preliminary Fatigue Analysis

Assumptions

• ADTT = 14,000; ADTTSL = 14,000×0.85=11,900

• AASHTO HS-15 Fatigue Truck (54 kips)• Category C to C’ detail; CAFT = 10 to 12 ksi• 2 lanes available to trucks• Stress Range = 14.2 ksi

1.0E+05 1.0E+06 1.0E+071

10

100

Category B Category C' Stiffeners Category C Other Attachments

Category D Cycle Limit with σ=14.2 ksi

N-Number of Cylces

Stre

ss R

ange

in k

si

AASHTO S-N Curve

14.2

1,537,000

Potential Repair Options

• Continue Monitoring / Drill Crack Arrestor Holes• Stiffen the Stiffener Plates• Remove retrofit links / diaphragms • Loosen diaphragm bolts / Install under-sized bolts• Soften the Stiffener Plates• Field Monitoring

Crack Arrestor Holes

AASHTO Fatigue Truck

Δσ = 5.7 ksi

Δσ = 13.0 ksi

Δσ = 25.2 ksi

AASHTO Fatigue Truck

27Positive Connection / Remove Links

Δσavg = 14.2 ksi

Δσavg = 0.7 ksi

Δσavg = 13.8 ksi

AASHTO Fatigue Truck

28Modify Connections / Loosen Bolts

Δσavg = 6.0 ksi

Δσavg = 14.2 ksi

Δσavg = 10.8 ksi

29

Average Stress (ksi)

Average Stress Reduction (ksi)

Crack Initiation (cycles)

Repair 1: Stiffen Plate (positive connection) 0.7 95% <CAFT

Repair 2: Remove Seismic Retrofit Links 13.8 3% 1.7×106

Repair 3: Loosen Bolts 10.8 24% 3.5×106

No Repair (Retrofit) 14.2 -- 1.5×106

Original Bridge 6.0 -- <CAFT

Repair Summary

Proposed Field Monitoring

• Validate Finite Element Analysis• Directly measure the stress-range cycles produced by random

variable live-load spectrum• Directly measure the number of cycles applied per interval of

time • Wireless monitoring of strain for approx. 1 month

Executive Summary

• Seismic retrofit resulted in unintended out-of-plane distortion• Cracks likely initiated within 1 year after the retrofit was

completed• At a minimum, continue drilling crack arrestor holes and

monitor for continued crack growth• Field monitoring (approx. 1 month) recommended• The key with any retrofit is to balance long-term durability

with seismic safety

Repair Option

Brief Description

Repair Methodology / Category

1

Remove the seismic retrofit links at bent 64 in an effort to increase flexibility.

Increase Flexibility

2

Loosen the bolts at the cross-frame to stiffener plate connection to allow rotation.

Increase Flexibility

3

Provide a positive connection between the stiffener plate and the top flange.

Increase Stiffness

4

Reduce the stiffness of the “×” cross frames by 50% by removing one of the angles from the double-angle section.

Increase Flexibility

5Continue to drill 2 ⅞-in. crack arrestor holes.

Increase Flexibility

6

Remove original “K” cross frames located at the center of the spans between bents 62 and 63 and bents 63 and 64.

Increase Flexibility

7

Replace the “×” braces added as part of the seismic retrofit with the original “K” braces.

Increase Flexibility

8

Increase the web-gap length by removing a portion of the connection plate.

Increase Flexibility