CHAPTER 2

G-2

&SPECIFICATIONS
BRIDGE FAILURES
Ad vis e r: E n g r. Ms . S a va n n a Bla ir S e rra o n
Prepared by: Christopher Salva BSCE -S E III

S ilve r Bridge , in P oint P le a s a nt WV

01
 BRIDGE
For mos t bridge e ngine e rs , it s e e ms tha t bridge s pe cifica tions we re a lwa ys the re . But tha t is not
the ca s e . The e a rly bridge s we re built unde r a de s ign–build type of contra ct. A bridge compa ny
would a gre e , for s ome lump-s um price , to cons truct a bridge conne cting one loca tion to
a nothe r. The re we re no s ta nda rd bridge s pe cifica tions a nd the contra ct we nt to the low bidde r.
The bridge compa ny ba s ica lly wrote its own s pe cifica tions whe n de s cribing the bridge it wa s
propos ing to build. As a re s ult, de pe nding on the inte grity, e duca tion, a nd e xpe rie nce of the
builde r, s ome ve ry good bridge s we re cons tructe d a nd a t the s a me time s ome ve ry poor
bridge s we re built.

Tra in b rid g e 1870s

02
 BRIDGE COLLAPSED
An event took place on the night of December 29, 1876,
tha t a ttra cte d the a tte ntion of not only the public but a ls o
the e ngine e ring profe s s ion. In a blinding s nows torm, a n 11-
ca r tra in with a double -he a de r locomotive s ta rte d a cros s the
As hta bula Cre e k a t As hta bula , Ohio, on a 175-ft (48-m) long
iron bridge ,whe n the firs t te nde r de ra ile d, plowe d up the tie s ,
a nd ca us e d the s e cond locomotive to s ma s h into the a butme nt
(Gie s , 1963).The coupling broke be twe e n the le a d te nde r a nd
the s e cond locomotive , a nd the firs t locomotive a nd te nde r
we nt ra cing a cros s the bridge . The bridge colla ps e d be hind
the m. The s e cond locomotive , te nde r, a nd 11 ca rs plunge d
s ome 70 ft (20 m) into the cre e k. The woode n ca rs burs t into
fla me s whe n the ir pot-be llie d s tove s we re ups e t, a nd a tota l
of 80 pa s s e nge rs a nd cre w die d.
03
 A S H T A B U LA H O W E T R U S S B R I D G E

In the investigation that followed, a number of shortcomings

in the wa y bridge s we re de s igne d, a pprove d, a nd built
we re a ppa re nt. An e xe cutive of the ra ilroa d who ha d limite d
bridge de s ign e xpe rie nce de s igne d the bridge . The a cce pta nce
of the bridge wa s by te s t loa ding with s ix locomotive s ,
which only prove d tha t the fa ctor of s a fe ty wa s a t le a s t 1.0
for tha t pa rticula r loa ding. The bridge wa s a Howe trus s with
ca s t-iron blocks for s e a ting the dia gona l compre s s ion
me mbe rs .
The s e blocks we re s us pe cte d of contributing to the fa ilure .
It is ironic tha t a t a me e ting of the Ame rica n S ocie ty
of Civil Engine e rs (AS CE), a s ta te me nt wa s ma de tha t “the
cons truction of the trus s viola te d e ve ry ca non of our s ta nda rd
pra ctice ” a t a time whe n the re we re no s ta nda rds of pra ctice

03
 BRIDGE SPECIFICATIONS
De s ig n
P h ilo s o p h y
1862- 1894 Lo a d a n d
Fa c to r De s ig n
Us in g c o n c e n tra te d a xle lo a d s (LFD) 1994 - 20 10 Lo a d a n d

01
in s te a d o f u n ifo rm ly d is trib u te d
lo a d s Ho we Tru s s 04 Firs t e d itio n o f th e AAS HTO
(1994) LRFD Brid g e De s ig n
Re s is ta n c e
Fa c to r De s ig n
b y Ch a rle s Hilto n o f th e Ne w Yo rk
Ce n tra l Ra ilro a d Brid g e S p e c ific a tio n s (LRFD)
(Ed wa rd s , 1959)

1894 - 190 3
P ro p o s e d h is o rig in a l c o n c e p t o f
02 tra in lo a d in g s with c o n c e n tra te d
a xle lo a d in g s fo r th e lo c o m o tive s
Th e Co o p e r s e rie s lo a d in g b e c a m e th e
s ta n d a rd in 1903 wh e n a d o p te d b y th e
a n d te n d e r fo llo we d b y a u n ifo rm ly
Am e ric a n Ra ilro a d En g in e e rin g
d is trib u te d lo a d re p re s e n tin g th e
As s o c ia tio n (AREA) a n d re m a in s in u s e to
tra in th e p re s e n t d a y.

1914 - 1921
by Theodore Cooper

03 Th e firs t e d itio n o f th e
S ta n d a rd S p e c ific a tio n s fo r
Hig h wa y Brid g e s a n d In c id e n ta l
S tru c tu re s wa s p u b lis h e d
b y AAS HO
THANK YOU!
EVERYONE!

10
 Silver Bridge,
Point Pleasant, UYANGUREN,
West Virginia, CHARLENE GRACE D.

December 15, 1967

 THE COLLAPSE OF THE SILVER
Silver Bridge, BRIDGE OVER THE OHIO
RIVER BETWEEN POINT
Point Pleasant, PLEASANT, WEST VIRGINIA,
AND KANAUGA, OHIO, ON
West Virginia, DECEMBER 15, 1967,
RESULTED IN 46 DEATHS, 9
INJURIES, AND 31 OF THE 37
VEHICLES ON THE BRIDGE
December 15, 1967 FELL WITH THE BRIDGE
(NTSB, 1970)
 SILVER BRIDGE WAS DESIGNED
BY J.E. GREINER COMPANY AND BUILT BY

Silver Bridge
GALLIA COUNTY OHIO RIVER BRIDGE
COMPANY (LATER WEST VIRGINIA-OHIO
BRIDGE COMPANY) AND ITS
SUBCONTRACTOR, U.S. STEEL’S AMERICAN
BRIDGE COMPANY. A TWO-LANE, 1760-FOOT-
LONG EYEBAR SUSPENSION BRIDGE WITH A
700-FOOT MAIN SPAN 102 FEET ABOVE THE
BOTTOM OF THE OHIO RIVER CHANNEL AND
TWO 380-FOOT ANCHOR SPANS, IT WAS
COMPLETED IN ONE YEAR, OPENING TO
TRAFFIC ON MEMORIAL DAY 1928.
Cause of Collapse
THE NATIONAL TRANSPORTATION
SAFETY BOARD (NTSB) FOUND
THAT THE CAUSE OF THE BRIDGE
COLLAPSE WAS A CLEAVAGE
FRACTURE IN THE EYE OF AN
EYEBAR OF THE NORTH
SUSPENSION CHAIN IN THE OHIO
SIDE SPAN (NTSB, 1970). THE
FRACTURE WAS CAUSED BY
DEVELOPMENT OF A FLAW DUE TO
STRESS CORROSION AND
CORROSION FATIGUE OVER THE
40-YEAR LIFE OF THE BRIDGE AS
THE PIN-CONNECTED JOINT
ADJUSTED ITS POSITION WITH
EACH PASSING VEHICLE
Effect on Bridge Practice
COLLAPSE OF THE SILVER BRIDGE
DISCLOSED THE LACK OF REGULAR
INSPECTIONS TO DETERMINE THE
CONDITION OF EXISTING BRIDGES.
CONSEQUENTLY, THE NATIONAL BRIDGE
INSPECTION STANDARDS (NBIS) WERE
ESTABLISHED UNDER THE 1968 FEDERAL AID
HIGHWAY ACT. THIS ACT REQUIRES THAT
ALL BRIDGES BUILT WITH FEDERAL MONIES
BE INSPECTED AT REGULAR INTERVALS NOT
TO EXCEED 2 YEARS. AS A RESULT, THE
STATE BRIDGE AGENCIES WERE REQUIRED
TO CATALOG ALL THEIR BRIDGES IN A
NATIONAL BRIDGE INVENTORY (NBI). THERE
ARE OVER 600,000 BRIDGES (100,000 ARE
CULVERTS) WITH SPANS GREATER THAN 20
FT (6 M) IN THE INVENTORY
 I-5 and I-210
BRIDGE TYPES IN THIS
Interchange, San INTERCHANGE INCLUDED
COMPOSITE STEEL
Fernando, GIRDERS, PRECAST
PRESTRESSED I-BEAM
California GIRDERS, AND
PRESTRESSED AND
NONPRESTRESSED CAST-
IN-PLACE REINFORCED
CONCRETE BOX-GIRDER
February 9, 1971 BRIDGES.
I-5 and I-210
Interchange AT 6:00 A.M. (PACIFIC STANDARD
TIME), ON FEBRUARY 9, 1971, AN
EARTHQUAKE WITH A RICHTER
MAGNITUDE OF 6.6 OCCURRED IN THE
NORTH SAN FERNANDO VALLEY AREA
OF LOS ANGELES. THE EARTHQUAKE
DAMAGED APPROXIMATELY 60
BRIDGES. OF THIS TOTAL,
APPROXIMATELY 10% COLLAPSED OR
WERE SO BADLY DAMAGED THAT
THEY HAD TO BE REMOVED AND
REPLACED (LEW ET AL., 1971). FOUR
OF THE COLLAPSED AND BADLY
DAMAGED BRIDGES WERE AT THE
INTERCHANGE OF THE GOLDEN STATE
FREEWAY (I-5) AND FOOTHILL
FREEWAY (I-210)
 MORE THAN ONE CAUSE CONTRIBUTED TO THE
COLLAPSE OF THE BRIDGES AT THE I-5 AND I-
Cause of Collapse 210 INTERCHANGE. THE BRIDGES WERE
DESIGNED FOR LATERAL SEISMIC FORCES OF
ABOUT 4% OF THE DEAD LOAD, WHICH IS
EQUIVALENT TO AN ACCELERATION OF 0.04G,
AND VERTICAL SEISMIC FORCES WERE NOT
CONSIDERED. FROM FIELD MEASUREMENTS
MADE DURING THE EARTHQUAKE, THE
ESTIMATED GROUND ACCELERATIONS AT THE
INTERCHANGE WERE FROM 0.33G TO 0.50G
LATERALLY AND FROM 0.17G TO 0.25G
VERTICALLY. THE SEISMIC FORCES WERE
LARGER THAN WHAT THE STRUCTURES WERE
DESIGNED FOR AND PLACED AN ENERGY
DEMAND ON THE STRUCTURES THAT COULD
NOT BE DISSIPATED IN THE COLUMN–GIRDER
AND COLUMN–FOOTING CONNECTIONS. THE
CONNECTIONS FAILED, RESULTING IN
DISPLACEMENTS THAT PRODUCED LARGE
SECONDARY EFFECTS, WHICH LED TO
PROGRESSIVE COLLAPSE. GIRDERS FELL OFF
THEIR SUPPORTS BECAUSE THE SEAT
DIMENSIONS WERE SMALLER THAN THE
EARTHQUAKE DISPLACEMENTS.
 POINTED OUT THE INADEQUACIES OF THE LATERAL
FORCE AND SEISMIC DESIGN PROVISIONS OF THE

Effect on Bridge Practice SPECIFICATIONS. MODIFICATIONS WERE MADE AND

NEW ARTICLES WERE WRITTEN TO COVER THE
OBSERVED DEFICIENCIES IN DESIGN AND
CONSTRUCTION PROCEDURES. THE ISSUES
ADDRESSED IN THE REVISIONS INCLUDED THE
FOLLOWING: (1) SEISMIC DESIGN FORCES INCLUDE A
FACTOR THAT EXPRESSES THE PROBABILITY OF
OCCURRENCE OF A HIGH-INTENSITY EARTHQUAKE
FOR A PARTICULAR GEOGRAPHIC REGION, A
FACTOR THAT REPRESENTS THE SOIL CONDITIONS, A
FACTOR THAT REFLECTS THE IMPORTANCE OF THE
STRUCTURE, AND A FACTOR THAT CONSIDERS THE
AMOUNT OF DUCTILITY AVAILABLE IN THE DESIGN;
(2) METHODS OF ANALYSIS CAPABLE OF
REPRESENTING HORIZONTAL CURVATURE, SKEWNESS
OF SPAN, VARIATION OF MASS, AND FOUNDATION
CONDITIONS; (3) PROVISION OF ALTERNATIVE LOAD
PATHS THROUGH STRUCTURAL REDUNDANCY OR
SEISMIC RESTRAINERS; (4) INCREASED WIDTHS ON
ABUTMENT PADS AND HINGE SUPPORTS; AND
(5) DISSIPATION OF SEISMIC ENERGY BY
DEVELOPMENT OF INCREASED DUCTILITY THROUGH
CLOSELY SPACED HOOPS OR SPIRALS, INCREASED
ANCHORAGE AND LAP SPLICE REQUIREMENTS, AND
RESTRICTIONS ON USE OF LARGE-DIAMETER
REINFORCING BARS.
Sunshine Skyway,
Tampa Bay,
Florida BERMOY, BETHANY

May 9, 1980
 Designed by Figg & Muller Engineering Group
The construction started June 1982
Opened April 20, 1987 (36 years ago)

Consists of a pair of long beam bridges with a

central cable-stayed bridge that spans lower
Sunshine Tampa bay to connect St. Petersburg, Florida,
to Terra Ceia.

Skyway Bridge The twin bridge structures are 4.24 miles (6.82
km) long and consist of posttensioned
concrete girder trestles, steel girder spans, steel
deck trusses, and a steel cantilever through
truss.
Sunshine Skyway
Bridge

9/6/2023 Implication of Bridge Failures on Practice 12

 Cause of Collapse

On the morning of May 9, 1980, the freighter MV

Summit Venture collided with a support pier near
the center of the bridge during a sudden squall.

The collapse of the cantilever through truss and

deck truss spans of the Sunshine Skyway Bridge was
due to the loss of support of the pier rammed by
the Summit Venture and the progressive instability
and twisting failure that followed.

Effect on Bridge Practice

The development of standards for the design,

performance, and location of structural bridge pier
protection systems. Provisions for determining vessel
collision forces on piers and bridges are now
incorporated in the AASHTO LRFD Bridge
Specifications.

Implication of Bridge Failures on Practice

NEW Sunshine Skyway
Bridge

9/6/2023 Implication of Bridge Failures on Practice 14

 The Mianus River Bridge is a steel deck bridge
of welded construction that has 24 spans, 19 of
which are approach spans, and is 2656 ft (810
m) long.

The highway is 6 lanes wide across the bridge,

but a lengthwise expansion joint on the

Mianus River centerline of the bridge separates the structure

into 2 parallel bridges that act independently
of each other.
Bridge
The deck structure over the river consists of two
parallel haunched steel girders with floor
beams that frame into the girders.
Mianus River
Bridge

9/6/2023 Implication of Bridge Failures on Practice 16

Mianus River
Bridge

9/6/2023 Implication of Bridge Failures on Practice 17

 Cause of Collapse
On the morning of June 28, 1983, the mianus river
bridge collapse because the failure of two pin and
hanger assemblies that held the deck in place on
the outer side of the bridge, according to an
investigation by the National Transportation Safety
Board.
Rainwater leaked down through the pin bearings,
causing them to rust. The outer bearings were
fracture-critical and non-redundant, a design flaw
of this particular type of structure. The bearings
were difficult to inspect close-up, although traces
of rust could be seen near the affected bearings.

Effect on Bridge Practice

The Mianus River Bridge was being inspected on a
regular basis, but the inspectors had no specific
directions as to what the critical elements were that
could result in a catastrophic failure.

The Mianus River Bridge, when the pin assembly

failed, it led to the collapse of the suspended span
because the structural system was non-load-path-
redundant.

Implication of Bridge Failures on Practice

Schoharie Creek
Bridge,
Amsterdam, New DERECHO, CARLOS
York,
April 5, 1987
The bridge design was approved in January
1952 by the New York State Department of
Transportation and began full service on
October 26, 1954. Its collapsed killed 10 people
and 5 vehicles fell in the river.

The bridge consisted of five simply supported

spans of lengths 30.5, 33.5, 36.6, 33.5, and 30.5
m. The roadway width was 34.3 m and carried
four lanes of highway traffic. The superstructure
was composed of two main steel girders 3.66 m
deep with transverse floor beams that spanned
the 17.4 m between girders and cantilevered
8.45 m on either side. Stringers ran
longitudinally between the floor beams and
supported a non-composite concrete deck
with members connected with rivets.
The substructure consisted of four piers and two
abutments. The reinforced concrete piers had
two columns directly under the two girders and
a tie beam near the top. A spread footing on
dense glacial deposits supported each pier.
The 2 middle piers in the main channel of
Schoharie Creek were protected by riprap and
Only the abutments were supported on piles.
Unfortunately, no reliable method was
available to predict scour depth when the
bridge was constructed.
Schoharie Creek highest water discharge was
2170 m3/s due to the flood on October 16,
1955. This caused slight damage on the riprap
and in 1977, a consulting engineering firm
recommended replacing the missing riprap but
was never done. The Schoharie Creek Bridge
inspected annually or biennially as required by
the National Bridge Inspection Standards of the
1968 Federal Aid Highway Act but only the
above water elements were inspected and
was done by maintenance personnel, not
engineers.
The severe flooding on April 5, 1987, with a
discharge of 1840 m3/s, caused local scour to
erode the soil beneath pier 3, which then
dropped into the scour hole, and resulted in
the collapse of spans 3 and 4. The bridge
wreckage in the creek redirected the water
flow so that the soil beneath pier 2 was eroded,
and some 90 min later it fell into the scour hole
and caused the collapse of span 2. Without
piles, the Schoharie Creek Bridge was
completely dependent on riprap to protect its
foundation against scour, and it was not there.

The NTSB determined that the probable cause

of the collapse of the Schoharie Creek Bridge
was the failure of the New York State Thruway
Authority to maintain adequate riprap around
the bridge piers, which led to the severe
erosion of soil beneath the spread footings.
Contributing to the severity of the accident
was the lack of structural redundancy in the
bridge.
The collapse of the Schoharie Creek Bridge
resulted in an increased research effort to
develop methods for estimating depth of scour
in a streambed around bridge piers and for
estimating size of riprap to resist a given
discharge rate or velocity. Methods for
predicting depth of scour are now available.
An ongoing problem that needs to be
corrected is the lack of qualified bridge
inspection personnel. This problem is especially
true for underwater inspections of bridge
foundations because there are approximately
300,000 bridges over water and 100,000 have
unknown foundation conditions.
Once again, the NTSB recommends that
bridge structures should be redundant and
have alternative load paths. Engineers should
finally be getting the message and realize that
continuity is one key to a successful bridge
project.
Cypress Viaduct,
Loma Prieta
DERECHO, CARLOS
Earthquake,
October 17, 1989
The California Department of Transportation
(Caltrans) has been a leader seismic design
and protection of bridges. They assess seismic
risk, update design procedures, and evaluate
existing bridges for catastrophic potential but
one of their main problems is gaining sufficient
funding.

The Loma Prieta earthquake occurred on

October 17, 1989, and it resulted in over $8
billion in damages and the loss of 62 lives. It
was one of the most reported-on structures by
the national media during that time as the
double-deck bridge failed in shear within the
columns and pancaked the bridge on traffic
below.
Caltrans was aware of the critical design
features that were necessary to provide the
ductility and energy absorption required to
prevent catastrophic failure. Unfortunately,
similar details were common in other bridge
substructures designed at the time. Caltrans
was working on correcting these defects, but
with over 13,000 bridges in its inventory and
limited resources, engineers had not been able
to retrofit the Cypress Viaduct before the
earthquake.

Because of this incident, the political will was

generated to significantly increase the funding
necessary to retrofit hundreds of bridges within
the Caltrans inventory. In addition, Caltrans
substantially increased its research efforts that
has resulted in many of the design
specification and construction details used
today.