19
Performance of Horizontally Curved Highway Bridge with Seismic Isolation Presented by: Ahmed Abdel-Mohti Authors: Eric V. Monzon, California State University – Sacramento Ian G. Buckle, University of Nevada – Reno Ahmad M. Itani, University of Nevada – Reno 1

Performance of Horizontally Curved Highway Bridge with ... › portals › 0 › files › pdf › webinars › ws_2018__… · Curved Bridge Response Experimental Results Isolator

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

  • Performance of Horizontally Curved Highway Bridge with Seismic Isolation

    Presented by: Ahmed Abdel-Mohti

    Authors: Eric V. Monzon, California State University – SacramentoIan G. Buckle, University of Nevada – Reno Ahmad M. Itani, University of Nevada – Reno

    1

  • Outline

    FHWA Curved Bridge Studies at UNR

    Seismically Isolated Curved Bridge

    ▪ Curved Bridge Response

    ▪ Experimental Results

    ▪ Isolator Instability

    ▪ Conclusions

    2

  • Bridge Geometry

    Prototype Model

    Total length 362.5 ft 145 ft

    Span lengths 105 ft – 152.5 ft – 105 ft 42 ft – 61 ft – 42 ft

    Centerline radius

    200 ft 80 ft

    Total width 30 ft 12 ft

    Girder spacing 11.25 ft 4.5 ft

    Column height 19.17 ft 7.67 ft

    Column diameter

    5 ft 2 ft3

  • Curved Bridge Model Geometry

    R = 80 ft at bridge c.l.Subtended angle = 104o (1.8 rad)

    Abutment 1(A1)

    Bent 2(B2)

    Bent 3(B3)

    Abutment 4(A4)

    4

  • L = 145 ft

    12 ft

    22 ft

    Total Model Weight = 167 kipsSuperstructure = 107 kipsSubstructure = 60 kips

    Weight of model:

    5

  • Experiments

    1. Conventional bridge (1040, steel bearings, 24 inch diam. columns, sacrificial shear keys at abutments)

    2. Conventional bridge with live load (6 trucks)

    3. Fully isolated bridge with 12 LRB isolators

    4. Hybrid isolated bridge with 6 LRB isolators and ductile cross frames

    5. Abutment pounding (nonlinear backfill)

    6. Rocking columns

    6

  • Construction of Bridge Model

    7

  • Seismic Isolators

    Abutment isolator

    Lead-rubber isolator

    8

  • Lead Rubber Bearings

    Parameters Abutment LRB Pier LRB

    Shear modulus, G 55 psi 55 psi

    Bonded diameter, B 7.5 in. 9.0 in.

    Layer thickness, tr 0.25 in. 0.25 in.

    Total rubber thickness, Tr

    2.75 in. 2.75 in.

    Total Height, H 7.0 in. 7.0 in.

    Lead core diameter, dL

    1.25 in. 1.5 in.

    Stiffness, Kd 0.86 kip/in 1.24 kip/in

    Characteristic strength, Qd

    1.41 kip 2.03 kip

    9

  • Performance of Column in Seismically Isolated Bridge

    0

    50

    100

    150

    200

    250

    300

    350

    400

    450

    500

    0.0000 0.0005 0.0010 0.0015 0.0020 0.0025 0.0030

    Res

    ult

    ant

    Mo

    men

    t (k

    ip-f

    t)

    Resultant Curvature (rad/in)

    Bent 2 Moment-Curvature

    SYL

    ELC

    HACH

    Section_Analysis_M-Phi

    Idealized_M-Phi

    10

    3 x Design EQ

    Design EQ

  • Superstructure Displacement

    0

    25.4

    50.8

    76.2

    101.6

    127

    152.4

    0

    1

    2

    3

    4

    5

    6

    0 25 50 75 100 125 150

    Dis

    pla

    cem

    ent

    (mm

    )

    Dis

    pla

    cem

    ent

    (in

    )

    % Design EQ

    Bridge with Seismic Isolation

    Bridge with Conventional Details

    Due to damage in the columns, in addition to effects of in-plane torsional rotation of the curved bridge, the conventional bridge experienced higher displacement at EQ levels beyond the Design EQ.

  • Column Shear Forces

    0

    10

    20

    30

    40

    50

    60

    0 25 50 75 100 125 150

    She

    ar (

    kip

    s)

    % Design EQ

    Total Resultant Shear: Pier 2

    0

    10

    20

    30

    40

    50

    60

    0 25 50 75 100 125 150

    % Design EQ

    Total Resultant Shear: Pier 3

    Conventional Bridge

    With seismic isolation

    Conventional Bridge

    With seismic isolation

    The seismic isolation reduced the column base shear by about 50% at Design EQ level.

    Due to bridge curvature, the supports experienced unequal base shears resulting in more damage at one support and less damage at the other. This, in turn, resulted in unequal stiffness of the columns which further increases the effect of in-plane torsional rotation.

  • Column Damage at 1.5 Design EQ (MCE)

    Pier 2Bridge with

    Seismic Isolation

    Pier 3Bridge with

    Conventional Details

    Pier 3Bridge with

    Seismic Isolation

    Pier 2Bridge with

    Conventional Details

  • Column Damage at 3x Design EQ

    Pier 2Bridge with

    Seismic Isolation

    Pier 3Bridge with

    Conventional Details

    Pier 3Bridge with

    Seismic Isolation

    Pier 2Bridge with

    Conventional Details

  • Isolator Deformation during instability

    21

    3

    456

    78

    9

    1011

    12

    15

    At 3x Design EQ, due to excessive displacements, instability occurred at the isolators at one of the abutments.

    However, global instability did not occur because the isolators at other supports remained stable. This, in addition to inertial loading in the bridge, the unstable isolators were able to return to original position at the end of the excitation.

    No residual displacement was observed.

  • Isolator Instability: 300% DE SYL

    16

    Video in real time

  • Abutment Isolator at 300% Design Earthquake

    17

    Video in Slow-Motion

  • Concluding Remarks

    • Essentially elastic behavior of columns can be achieved in a seismically isolated bridge even at earthquake levels equal to three times the design.

    • In a continuous curved bridge with multiple spans, instability of isolators at one or two supports does not lead to bridge collapse.

    • Lead-rubber isolators are highly resilient seismic protective systems. Extreme deformation and instability in these isolators has minimal effect on their stiffness.

    18

  • 19