Chesapeake City Bridge Crack Study

Chesapeake City Bridge Crack Study

• Adrian Kollias, P.E.• Philadelphia District Bridge Program

Manager

US Army Corpsof EngineersPhiladelphia District

Introduction

Overview

• Present problem• Previous repair attempts• Modeling • Final solution

MARYLAND

DELAWARE

NEW JERSEY

301

13

40

1

95

95

295

Baltimore

Philadelphia

Wilmington

C&D Canal

DoverD

EL

AWA

RE

MA

RY

LA

ND

ChesapeakeBay

DelawareBay

Chesapeake & Delaware Canal Crossings

Maryland Delaware

ChesapeakeCity Bridge

2 Lanes

SummitBridge4 Lanes

St. GeorgesBridge4 Lanes

Reedy PointBridge2 Lanes N

Conrail

DERteUS

Rte

MDRte

DE Rts80671US 9

13213

301

Terminology• Fracture Critical Members: tension members

or tension components of members whose failure would be expected to result in the collapse of partial collapse of a bridge

• Fatigue: the tendency of a member to fail at a lower stress when subjected to cyclical loading than when subjected to static loading.

• Fatigue crack – any crack caused by repeated cycle loading.

• Fatigue life – the length of service of a member.

Chesapeake City Bridge

Tie GirderFloorbeam

Arch

Pier

– Tied-arch structure– Two traffic lanes, Maryland Rte. 213– 3,954 feet in length– Two-girder, fracture critical structure– ADT = 14,825 (2004)– ADTT = 2,635 (2006)– Constructed 1947-1948– Overall structural condition is fair– Design live load: HS20-44

Description

Cracks at 3 Locations: L0, L0’, L1’

Bridge Floor System

Sliding Bearings

Cracked Connection Angle Locations

Stringers

Floorbeam

Deck

Tie Girder

Crack Location

Track Crack Propagation with Bi-weekly Inspections

Crack Location

Track Crack Propagation with Bi-weekly Inspections

Crack Location

Track Crack Propagation with Bi-weekly Inspections

Chesapeake City Bridge

Reason for Concern– Public Safety– Potential for partial bridge failure if

corrective measures are not taken– Major traffic thoroughfare connecting both

northern and southern Delmarva Peninsula in Maryland

– Connects Northern and Southern Chesapeake City

Attempt #1Drilling Holes

Attempt #2Replace Top Portion of Cracked Angles

New Angle Section

After failed Attempt #2, developed numerical models to

investigate the cracking and analyze bridge behavior.

Determine that frozen stringer bearings are causing the cracks and must be

replaced.

Original Bronze Bearings

Stringer

Floorbeam Top Flange

Sole Plate

Bearing Plate

Bronze Plate

Filler Plate

Original Bronze Bearings

Crevice Corrosion

Attempt #3: Replace “Frozen” Stringer Bearings

StringerSliding Bearings

Floorbeam

Diaphragm

New Neoprene BearingsSole Plate

Bearing PlateNeoprene Bearing Pad

• Repairs performed in 2003- Replaced 72 bearings out of 180 total- Repaired connection angles for 6 floorbeams out of a possible 16 total

• Cost: $945,000• Duration: 210 calendar days

• Cracks reappear at the angle connections 1-year after bearing repair.

• Need to re-evaluate numerical models and design a repair retrofit for the angles to prevent future cracking.

Global Model

Load 1X

Y

Z

Global Modeling: Details and Assumptions

– Modeled using STAAD.Pro 2005– Created using beam and shell elements– All members modeled as beam, except deck slab which is

modeled using shell elements– Rigid elements and offsets to account for differences in c.g.

locations of members– New elastomeric stringer bearings modeled as tri-directional

linear springs– Remaining original stringer bearings are modeled as restrained

in 3 directions– South main arch bearings free to expand longitudinally and

rotate about transverse axis– North main arch bearings fully fixed– Deck is continuous (i.e., can transfer axial force from one panel

to another)

Calibration of the Global Model– Calibrated to measured global deflection data– Calibrated to measured strains from two

previous diagnostic tests– Overall goal of the calibration

• Capture the key features of the global response in terms of global deflection and floorbeam stress

• Strive for realistic agreement in magnitudes, given very complex behaviors and small magnitudes of measured deflection and stress

Initial Findings

a. Cracking is Due to Relative Rotation between Tie Girder & Floorbeam

b. Cracking is Due to Fatigue not Strength

b. Continuous Deck Model Best Predicts Floorbeam Stresses Matching Actual Field Measurements

c. Frozen Stringer Bearings and Stiff Deck Joints are both Contributing to the Cracking

Deflection Under Test Vehicle

Model Results Measured Stress in Top Flange - Floorbeam L1'

-3.0

-2.0

-1.0

0.0

1.0

2.0

0 100 200 300 400 500

Load Position (ft)

Stre

ss (k

si)

Gage 13 MeasuredGage 16 MeasuredGage 16 Model 6Gage 13 Model 6

Measured Stress in Top Flange - Floorbeam L1'

-3.0

-2.0

-1.0

0.0

1.0

2.0

0 100 200 300 400 500

Load Position (ft)

Stre

ss (k

si)

Gage 13 MeasuredGage 16 MeasuredGage 13 Model 8Gage 16 Model 8

Discontinuous Slightly ContinuousMeasured vs. Predicted Stress in Top Flange - Floorbeam L3'

-3.0

-2.0

-1.0

0.0

1.0

2.0

0 100 200 300 400 500

Load (Rear Axle) Position (ft)

Stre

ss (k

si)

Gage 28 MeasuredGage 27 MeasuredGage 28 Model 18Gage 27 Model 18

Completely Continuous

Remove Sample of Rubber Deck Joint Material to Test

Stiffness

Deck Joints

Deck Joint

Original Deck Joint Design - 1977

Rubber Seal

½” x ¼” Steel Support Bars

Fused Steel Bars

Deck Joint

Deck Joints are Restrained from Movement

Fused Steel Bars

Typical Deck Joint

Fused Steel Bars

Typical Deck Joint

Joint Busters IDouble Click to See Video

Joint Busters IIDouble Click to See Video

Models indicate existing FTGC anglesdo not achieve infinite fatigue life even with bearings and deck joints repaired.

Retrofit Design Process• Obtain Design Forces – Global Model

• Develop Preliminary Retrofit Designs (2 Stiffened + 2 Softened)

• Incorporate Retrofit – Local Model

• Verify Retrofit Effects - Global Model

• Finalize Retrofit Design

Local Model

Fatigue Analysis

• Fatigue life is function of stress range• Conducted using actual traffic data

(cycles) and vehicle weight crossing bridge

• Fatigue category C for out-of-plane displacement behavior

• Criteria from AASHTO Guide Specifications and LRFD Specifications

Current Repair Contract • Replace top portions of FTGC angles with

thicker angle members at L0 to L5 and L1’ to L5’.

• Replace all deck joint compression seals• Replace neoprene bearings at exterior stringers

at Floorbeams L1 to L3 and L1’ to L3’.• Restore bronze plate bearings at Floorbeams L4

to L7 and L4’ to L7’.• Cost: $1.3 million

Questions?