Performance Evaluation of Longitudinal UHPC Closure Pour Connection for Use in Modular Bridge Construction ―― A Pairwise Comparison of Capacity and Ductility at Failure Limit State Yaohua Deng (Jimmy) Brent M. Phares Andrew J. Putz

Curtis Carter Michael Nop Dean Bierwagen

Bridge Engineering Center Iowa State University

Office of Bridges and Structures Iowa Department of Transportation

Acknowledgment o Iowa Department of Transportation (IADOT) o Federal Highway Administration (FHWA)

Background and Motivation o ASCE 2013 infrastructure report card  United States: aging infrastructure including 600,000+ bridges  > 11% (↑): structurally deficient  Bridges approaching the completion of design life: replacement I-5 bridge over the Skagit River in Washington (from Google)

I-35W Mississippi River bridge (from Google)

Background and Motivation • Why rapid renewal techniques? o increased traffic congestion o economic constraints o work-zone safety concerns

• Why Accelerated Bridge Construction (ABC) techniques? o effective rapid renewal techniques (SHRP 2) o commonly utilizing prefabricated bridge elements connected with both transverse and longitudinal joints using high performance materials o Keg Creek Bridge (previous) and Little Silver Creek Bridge (current)

Background and Motivation • Little Silver Creek Bridge

Longitudinal joint

Transverse joint

Longitudinal joint

Prefabricated deck unit

Background and Motivation • Longitudinal joints o Important deck-level component  transferring shear and moment between these prefabricated components.  heavily stressed by traffic loads and environmental effects o Three types of longitudinal connections  welded steel connection (low flexural resistance/cracking/leakage)  distributed reinforcement connection (constructability and bridge applications)  ultra-high performance concrete (UHPC) connection (reduced joint width/simplified reinforcement configuration)

Objectives To evaluate the flexural behavior and strength and failure modes of the longitudinal UHPC connections through: •

Experimental Testing  ponding and strength tests  comparison between a UHPC connection and a nonjointed detail

• Finite Element Analysis

Experimental Program

Specimen Design Longitudinal connection

Specimen design

Prefabricated deck unit

Type of Slab Specimens No. C1 Jointless C2 Specimens C3 J1 Jointed Specimens J4 J2

Joint Surface Preparation Technique N/A N/A N/A Rubber sandblast medium Plastic sandblast Retarder

Specimen Design – Jointed Specimens

Longitudinal Joint Detail

6" Bar Stagger Between Modules

10" #7 Bars

5''

5''

UHPC

J1

8'' #6 Bars

Jointed Specimens J4 J2

Rubber sandblast medium Plastic sandblast Retarder

Specimen Design – Jointless Specimens

“Connection location”

C1 Jointless C2 Specimens C3

N/A N/A N/A R bb

dbl

Specimen Fabrication

Ponding and Strength Tests Pond

1'-5.00''

4'-4.00''

3'-4.00"

4'-4.00''

7'

10.00"

7'

3

4

3

Strain gage

1 Deflection transducer

2

1

2 Joint line location for jointed specimens

1'-5.00''

4

Strain gage Deflection transducer Joint line location for jointed specimens

Finite Element Analysis

FE Model of jointed (or jointless) specimens

Line Loads Top Line Restraint

Connection

Deck Panel

Line Support

Concrete: Eight-noded solid elements Steel bars: Link elements

FE Model of jointed (or jointless) specimens Stress-strain relationships: • •

  ε   ε 2  fc = f c'  2   −    HPC   ε o.n   ε o.n   (Hognestad):   ε n  UHPC fc = fc' 1 − 1 −     ε o.u   (Schmidt and

Tensile strengths: ε o.n = 0.00048( f c' )1/4

•

ft ' = 0.24 fc'

(AASHTO):

ε o.u

f c' = Ec.u

Ec.u = 1550 f

' c

Fehling):

Perfectly elastic–plastic uniaxial material model for steel • Yield strength 60 ksi • Elastic modulus 29000 ksi • Poisson’s ratio: 0.3

HPC

• UHPC (Russell and Graybeal)

ft ' = 0.21 fc'

Results and Discussions

Ponding Tests Pond

Curing of the UHPC  no cracks were visually found During the ponding tests  no leakage was found Conclusions:  A good bond at the interface  No cracks due to early-age drying shrinkage and temperature change

Strength Tests – Cracking Patterns and Failure Modes Cracks

Cracks

Flexural-shear failure Concrete crushing

Cracks

Flexural-shear failure Concrete crushing

Cracks

Strength Tests – Loads, Strains, and Deflections Jointless specimen-C1

Jointed specimen-J2

Plateau

Plateau

3 1 4

3

Strain gage Deflection transducer

2

1

2 Joint line location for jointed specimens

4

Strain gage Deflection transducer Joint line location for jointed specimens

Strength Tests – Loads and Deflections

Type of Slab Specimens No. C1 Jointless C2 Specimens C3 J1 Jointed Specimens J4 J2

Joint Surface Preparation Technique N/A N/A N/A Rubber sandblast medium Plastic sandblast Retarder

Load at Concrete Cracking (kips) Test FEA 52 57 56 39 34 20 31

Load at Steel Yield (kips) Test FEA 115 123 164 126

112 114

Test 250 235 230 195

110 49

Deflection at Maximum Load (in.) FEA STM Test FEA 0.37 232 225 0.42 0.27 0.39

Load at Specimen Failure (kips)

180

225 210

--210

205

0.39 0.38

0.16

Strength Tests – Strut-And-Tie Models

Lateral restraining action

Arching action

Arching action

Lateral restraining action

Top layer

Steel yield

Bottom layer

Steel yield

Bottom layer

Conclusions o The UHPC connections had no cracks or leakage due to early-age drying shrinkage and temperature changes. o The jointed specimens had slightly lower cracking loads compared to those of the jointless specimens due to lower bond. A flexural-shear failure mode was found in the jointless and jointed specimens. Cracks formed at the connection interface and no cracks or concrete crushing were found in the UHPC pour. The strength and ductility of the jointed specimens are comparable to those of the jointless specimens. o Using the developed FE models, the failure loads were accurately predicted but the cracking and yield loads were over-estimated and the deflections at failure were underestimated. Using the developed STM model the ultimate load capacity of the specimens were accurately estimated.

Questions?