Why Buildings Fail - Washington Association of Building Officials

Why Buildings Fail
Norbert Delatte, P.E., Ph.D., F.ACI, F.ASCE
Professor and Chair
Department of Civil and Environmental Engineering
Cleveland State University
Structural Engineering
 “The art of moulding materials we do not
Why Buildings Fail
Norbert Delatte, P.E., Ph.D., F.ACI, F.ASCE
Professor and Chair
Department of Civil and Environmental Engineering
Cleveland State University
really understand into shapes we cannot
really analyze, so as to withstand forces
we cannot really assess, such a way that
the public does not really suspect.”
 E. H. Brown
 Also applied to geotechnical engineering,
etc.
Failure
 An unacceptable difference between
expected and observed performance
 Involves functional failures as well as
collapses
Failure Literacy
 Knowledge of well known failure case




Engineering Design
 Selecting materials, systems, and
configurations in order to ensure
satisfactory performance over the
projected lifetime of the facility
histories
Understanding of causes behind failures
Recognition of patterns of failures
Understanding of role of failure in
development of design procedures and
codes
Skepticism
Delatte’s Simplified Definition
 Anticipate everything that can possibly go
wrong
 Make sure it doesn’t happen
1
Categories
 Collapse During Construction
 Collapse in Service Related to Design and
Construction Errors
 Collapse in Service Due to Unsafe
Modification
Collapse in Service Related to
Design and Construction Errors
 Ronan Point
 Hartford Civic Center
 Kemper Arena
 Citicorp Tower (did not collapse)
Collapse During Construction
 2000 Commonwealth Avenue
 Skyline Plaza/Bailey’s Crossroads
 Harbour Cay Condominium
 L’Ambiance Plaza
Collapse in Service Due to
Unsafe Modification
 Sampoong Super Store
 Hotel Vendome
Beyond Failure – all proceeds to
the Isabella and Joe Delatte
education fund and Norb and
Lynn Delatte second honeymoon
fund
2
Collapse of 2000
Commonwealth Avenue
Suzanne King,
Roger Williams University
Faculty Advisor: Dr. Norbert Delatte
Introduction
2000 Commonwealth Ave. collapse
occurred on January 25, 1971
Many factors contributed to
collapse
Construction
Outline
Introduction
Building
description
Construction
Collapse
Causes
Other cases
Lessons
learned
Conclusions
Building Description
Sixteen story high-rise apartment
building
Cast-in-place reinforced concrete
Flat slab construction
Central elevator shaft
Penthouse mechanical room with a
five-foot crawl space
Floor Plan
Excavation began in fall of 1969
Most of work was subcontracted
At time of collapse, construction
was almost complete
Brickwork up to 16th floor and
work had started in individual
apartments
1
Collapse
Phase 1: Punching Shear Failure in
the Main Roof at Column E5
Phase 2: Collapse of Roof Slab
Phase 3: General Collapse
Phase 2: Collapse of Roof Slab
After hearing a warning, most workers
managed to get out of the way
Roof slab began to form the shape of a
belly
Roof collapsed onto sixteenth floor
Reinforcing steel was being placed, so
workers were forced to cross over to
the west side of the building
Phase 1: Punching Shear
About 3:00pm workers take break
from placing concrete for the
mechanical room floor slab
Placement started at the west edge
and proceeded east
Shortly after break there was a drop
in the floor slab
Punching shear was noticed around
column E5
Phase 3: General Collapse
Progressive collapse occurred 20
minutes after roof collapsed
Weight of the roof caused the 16th
floor to collapse onto 15th and so on
down to the ground
Two thirds of the building was gone
Four workers died
Extent of Collapse
Boston
Globe
Photos
2
Causes of Failure
Punching shear failure at column
E5
Design Flaws
Procedural/Construction flaws
Punching Shear
What is Punching Shear?
Happens when floor slab becomes
too heavy for a column to hold
Added construction loads can cause
unbalanced moments
The dead weight the column cannot
hold is transferred to surrounding
columns. However, this added
weight is often too great and the
floor slab fails.
Causes of Punching Shear
Concrete strength was well below
required 3000 psi
Inadequate shoring under the roof
slab
Construction equipment and two
boilers were on the roof
Design/Construction Flaws
Insufficient length of rebar

the bars did not extend enough into
columns
Incorrect placement of bars



confusion with deliveries
design around columns did not meet ACI
codes
billet steel vs. rail steel
3
Procedural/Construction
Flaws
Lack of proper building permit and
field inspection
Premature removal of formwork
Lack of construction control
Confusion Surrounding
Construction
8-27-69
8-29-69
9-4-69
9-5-69
fall ‘69
fall ‘69-70
11-10-70
new permit, ownership change
affidavit states plans meet code
change in ownership
permit granted for 16 story
building
construction begins
concept changes
ownership changes
1-25-71
building collapses
Confusion Surrounding
Construction
11-3-64
5-24-65
8-16-67
11-20-67
7-3-68
12-23-68
7-7-69
8-1-69
Similarities to Other Failures
Skyline Plaza in Bailey’s Crossroads, VA


Twenty eight day cylinder test
Slump test
Punching shear
What happens when shoring and
formwork are not properly used
March 2, 1973 collapse while under
construction
premature removal of shoring and
insufficient concrete strength
Harbour Cay Condominium Building, Cocoa
Beach, FL


Lessons Learned
first building permit applied for
excavation begins
permit lapsed due to delay
new permit, new building design
zoning change obtained
ownership changes
new permit, new building design
excavation continued
March 27, 1981 collapse while under
construction
punching shear failure
Conclusion
The collapse of 2000 Commonwealth
Avenue could have been avoided
It is the responsibility of everyone
involved in a project to see that all
codes and design specifications are
being adhered to
4
Key Reference
City of Boston report
Acknowledgements
Case study developed using support
provided by the National Science
Foundation under the project
“Developing Case Studies in Failures
and Ethics for Engineering Educators,”
project number DUE 0127419, Program
Director Robert W. Simoneau
5
Bailey’s Crossroads, VA, 1973
 March 2, 1973 collapse while under
Bailey’s Crossroads,
Virginia, 1973
construction – 14 killed, 34 injured
 Just over 2 years after 2000
Commonwealth Ave
 Investigated by National Bureau of
Standards at OHSA’s request
New Reference: Jeffrey Schellhammer; Norbert J. Delatte,
and Paul Bosela, “Another Look at the Collapse of Skyline
Plaza at Bailey's Crossroads, Virginia,” ASCE Journal of
Performance of Constructed Facilities, May/June 2013.
Project Background
Project Background
 26-story apartment building under
construction, part of $ 200 million
residential program
 Designed to ACI 318-63 standard
 Construction at a pace of one floor per
week
 At end of February, construction pace
increased to nearly twice that rate
Flat Plate Construction
 Slab carries shear and bending forces and
stresses without reinforcement from
beams, drop panels, capitals
 Efficient form of construction
 Bottom of slab serves as ceiling for apartment
below
 Allows for short story height, more floors per
building
 Prone to punching shear failure
1
The Collapse
 Shortly after lunch, March 2, 1973
 Workers observed slab deflections 6




inches to 2 feet at 23rd and 24th floor slabs
Section of building collapsed to the ground
Impact caused adjacent parking garage to
collapse
14 workers killed and 34 injured
Initial hypothesis – caused by falling crane
National Bureau of Standards
 National Bureau of Standards (NBS), now
National Institute of Standards and
Technology (NIST) asked to investigate by
OSHA
 Arrived on March 5, eastern part of the
building had already been demolished
NBS Scope of Investigation
 Cause of the accident
 Whether there had been any violations of
OSHA safety standards
 If any violations contributed to the collapse
2
Formwork and Shoring
Shoring Requirements
 NBS needed to determine state of
 Engineer’s structural drawings required 2
formwork, shoring, and reshoring was at
time of collapse
 Some reshoring had been placed to
stabilize building after collapse
 Worker accounts conflicted
levels of shores + 1 level of reshores
 OSHA required shoring left in place for 10
days of temps over 50°F
Construction Status
Concrete Strength Development
 Maturity concept M(t) = Σ(Ta – T0) ∆t
 Equivalent age
 Translation – concrete gains strength
based on both time and temperature
 Low temperatures → need more time
before forms may be removed
Analysis
 February temperature range 28 – 52° F
 Estimated compressive strength only 960
– 1,440 psi at time of collapse
 Early use of computer finite element
analysis
 Flexural failure unlikely
 Punching shear likely given concrete
compressive strength
3
Legal and Ethical Implications
 Construction halted
 Subcontractor fined less than $ 20,000 for
improper shoring
 Vice President of concrete firm tried for
manslaughter, hung jury
 Worker’s Compensation legislation
provided immunity to contractors and subs
Recommendations – Dov
Kaminetzky
 Contractor must prepare formwork,




shoring, and reshoring drawings
Contractor needs detailed plan for
stripping forms, including cylinder tests
Inspectors and agencies need to verify
that contractor performs these two tasks
Engineer of Record provides design loads
Use continuous reinforcement
Legal and Ethical Implications
 Structural engineers and architects held
responsible, although design was
adequate and collapse was due to poor
construction practice
 Design requirements not followed on site
 Sued for $ 500,000, which was later cut in
half
Conclusions
 Premature removal of shoring and
insufficient concrete strength
 Relatively ordinary building
 Fairfax, Virginia adopted a formal critical
structures program, requires
preconstruction conferences to set
responsibility
Photo Credits
 Photos provided by Nick Carino,
NBS/NIST
4
Introduction
L’Ambiance Plaza Case Study
Case Study by Rachel Martin and
Norbert Delatte
28 workers died during construction
collapse
April 23, 1987, approximately 1:30 p.m.
Many theories developed to explain
collapse
Workshop July, 12, 2003 at the University of Alabama
at Birmingham – funded by the National Science
Foundation (DUE 0127419), with support from the
ASCE Technical Council on Forensic Engineering
Education Committee
Design and Construction
Floor Plan of L'Ambiance Plaza
16-story building with 13 apartment
levels topping 3 parking levels
Two offset rectangular towers
Constructed using lift-slab system


Floor slabs for all 16 levels cast on ground
Packages of 2 or 3 slabs jacked into
position, then fixed in place
System patented by Youtz and Slick,
1948
Collapse
Status of Construction at Time
of Collapse
Workers tack welding wedges to hold
up 9th, 10th, 11th floor slab package
Loud metallic sound followed by
rumbling
Slab “cracked like ice breaking”
Structure collapsed in 5 seconds
1
Ruins of L’Ambiance Plaza
Cause of Failure
Six competing theories developed by
different investigators
Image from http://www.sgh.com/aplaza.htm
Theory 1 – Overloaded Steel
Angle Deformed
Lifting Assembly – L’Ambiance
Plaza
National Bureau of Standards (NBS),
now National Institute of Standards and
Technology (NIST)
Failure occurred at building’s most
heavily loaded column, E4.8 or E3.8
Lifting assembly failure
Angle deformed, rods slipped out
Remark in paper that theory was
withdrawn is incorrect, see discussion
Failure Sequence
Theory 2 – Unstable Wedges
Thornton-Tomasetti
Instability of wedges at column 3E
caused 12 floor and roof packages to
fall
Very high gaps at shearheads
Lateral loading allowed wedges to roll
out
2
Wedged Slab-to-Column
Connection
Theory 3 – Improper
Posttensioning Tendon Design
Schupack Suarez
In West Tower, tendons split apart at
column 4.8E
No tendons to carry force through
column line
Design details did not account for slab
openings
General Layout of
Posttensioning Tendons
Theory 4 – Poor Weld Details
and Welds
Occupational Safety and Health
Administration (OSHA)
30 welds examined around shearheads,
only 13 found acceptable
One-sided square-groove welds were
not prequalified joints
Theory 5 – Global Instability
Failure Analysis Associates
Structure was laterally flexible during
lifting and welding operations
Jacking operations could have triggered
collapse
Theory 6 – Foundation Failure
Oswald Rendon-Herrero
Geotechnical conditions under structure
included disintegrated rock and fill of
varying quality, mica, and micaceous
schists
Foundation failure could have caused
critical column to slip
3
Legal Repercussions
All theories plausible, but what
triggered collapse?
Two-judge panel mediated settlement
to close case
20 or more parties found guilty of
“widespread negligence, carelessness,
sloppy practices, and complacency.”
$ 41 million settlement fund
Professional and Procedural
Aspects
Responsibility fragmented among many
subcontractors, design deficiencies not
detected
Need standardized procedures for liftslab construction
Licensed engineer should be present
during construction
Educational Aspects
Safety during construction
Load paths
Need for redundancy
Fix overall responsibility for project
Technical Aspects
During all stages of construction,
ensure lateral stability of structure
Check punching shear, connection
redundancies
Provide sway bracing during lifting
operations (required but not used at
L’Ambiance Plaza)
Ethical Aspects
Structural safety not adequate during
construction
Public safety includes safety of workers
Summary and Conclusions
Final determination of collapse
mechanism not made, but theories
published
Connecticut now requires peer review
of complicated projects
Collapse severely reduced the use of
the lift-slab technique
4
Cleveland Pigeonhole Parking
Garage
April 6, 1956
5
Cynthia Rouse
Design and Construction
Larsen-Nielson
System –
Denmark, 1948
Case study by
Cynthia Pearson and
Norb Delatte
Precast
Large Panel Concrete
22
floors high – 5 flats per
floor – total 110
Took
less than 2 – years to
complete (July 25, 1966-March
11, 1968)
Ronan Point
Apartment Tower
– Canning Town,
East London,
England –1968
Precast Column Large Panel Construction
Collapse
System Building
Southeast
corner
collapsed 5:45 a.m.
Need for new housing stock after WW
II
Shortage of skilled construction workers
New method needed for



by gas
explosion on 18th floor
All
exterior walls load
bearing
Speed of construction
Minimum building footprint (high-rise)
Simplified construction methods
Floor 18;
Apartment 90
Collapse
Collapsed
Causes of Failure
in 2
phases
Collapse
sheared
off living room
section of the
apartments
4
fatalities/17
injuries
Title goes here
Initiated
Substandard
brass nut
fractured
Relatively
small
explosion – less than 10
psi
Tests
show that walls
could be laterally
displaced at 2.8 psi
Lack
of structural
redundancy
1
Cynthia Rouse
Remedial Actions
Taken
Joint Details
Southeast corner was rebuilt
Building reinforced with blast angles
Gas banned from Ronan Point
Technical Concerns
Fire and/or strong wind could have
similar effect on building
Designed to withstand 63 mph
wind
105 mph wind at 200 ft. up was
expected to occur within life of
building
System was not intended for more
than 6 floors
Poor workmanship was found
Building
Demolition
May 1986
Architect Sam Webb insisted on
dismantling floor by floor
Allowed study of joints
Title goes here
Other Findings
Fire test confirmed poor fire safety
Joints opened in service, smoke and
small objects could pass
Survey found considerable cracking and
movement
Building
Demolition
Webb: “I knew we were going to find bad
workmanship – what surprised me was the
sheer scale of it. Not a single joint was correct.
Fixing straps were unattached: leveling nuts
were not wound down, causing a significant
loading to be transmitted via the bolts: panels
were placed on bolts instead of mortar. But the
biggest shock of all was the crucial H-2 loadbearing joints between floor and wall panels.
Some of the joints had less than fifty percent of
the mortar specified.” (Wearne, 2000).
2
Cynthia Rouse
Professional and
Procedural Aspects
Extent
of poor workmanship led to the
demolition of other Larsen-Nielson system
buildings.
Building
regulations were revised.
Engineering
community was reminded of
concept of redundancy.
Building Code Changes
Fifth amendment – 1970
Structure must remain stable (with
reduced safety factor) with member
removed
Must resist 51 psi pressure from any
direction
The
importance of quality control was
recognized.
Ethical Aspects
Information
was hidden from the public
because of political issues.
Government
hushed Sam Webb, threatened
with jail time.
Architect,
Sam Webb took a position to
protect public’s safety.
Educational
Lessons learned: Aspects
Importance
Importance
of building codes and construction
quality control.
Applications:
Calculations
Free
body diagrams
Challenge
Conclusions
Flawed in design and execution
Existing building codes inadequate for
this structure
Lacked alternate load paths to
redistribute forces
Poor workmanship at critical
connections
Title goes here
of continuity in structural design.
students to think
References
Pearson and Delatte, “Lessons from the
Progressive Collapse of the Ronan Point
Apartment Tower,” Proceedings of the
3rd ASCE Forensics Congress, 2003
Levy, Matthys and Salvadori, Mario,
(1992). Why Buildings Fall Down.
W.W. Norton and Company, New York,
New York.
3
Cynthia Rouse
References
Wearne, Phillip (2000). Collapse: When
Buildings Fall Down. TV Books, L.L.C.,
USA.
Title goes here
Acknowledgements
Case study developed using support
provided by the National Science
Foundation under the project
“Developing Case Studies in Failures
and Ethics for Engineering Educators,”
project number DUE 0127419, Program
Director Robert W. Simoneau
4
Hartford Civic Center Case
Study
Case Study by
Rachel Martin
and Norb Delatte
1
From LZA report
Design and Construction
1970 – Innovative design proposed for
300 by 360 foot space frame roof, 83
feet above arena
Two main layers in 30 by 30 foot grid,
21 feet apart
30 foot members braced by midpoint
diagonals
Unusual Elements
Hartford
Civic
Center
Roof
Compression Member
Configurations
Four angles arranged in cross, not I or
tube shape – low buckling strength, but
easy to connect
Top horizontal members did not directly
intersect diagonals
Roofing panels on top of short posts
Supported on pylon legs
Space frame not cambered
Use of Computers
Too complex for hand analysis
Early use of “state-of-the-art” computer
analysis
Deflection predictions – 13 inches down
at center, 6 inches up at corners
Construction
Roof assembled on the ground
Excessive deflections found at some
nodes – engineers notified
Roof jacked up onto pylons –
deflections twice as much as predicted
Subcontractor had difficulties attaching
fascia panels
Pattern – large, unanticipated
deflections
2
3
Collapse
Roof collapsed under snow load at 4:19
a.m. on January 18, 1978
Evening before, arena had been packed
for a basketball game
No loss of life, no injuries
4
Investigation
Hartford appointed 3 member panel
Lev Zetlin Associates, Inc. (LZA)
LZA issued report later that year
Cause of Failure
Roof began failing as soon as
completed, due to design errors
Dead loads underestimated by 20 %
Three design errors – top layer
compression members



Some overloaded by 852 %
Other members overloaded by 213 %
Interior members overloaded by 72 %
5
Bracing,
Buckled
Shapes,
and
Capacities
From LZA report
Connection A
Connection B
Original
As-built
Original
As-built
Capacity 160,000 lb.
Capacity 15,440 lb.
Capacity 185,000 lb.
Capacity 59,000 lb.
Connection C
Other Findings
Original
As-built
Capacity 625,000 lb.
Capacity 363,000 lb.
Roof posts also applied bending stress,
not considered in design
Spacer plates for built up members too
far apart – violated AISC slenderness
ratio
Some members had bolt holes > 85 %
of cross section, violated AISC
Some misplaced diagonal members
6
Other Investigations
Loomis and Loomis, Inc. – blamed
torsional buckling, not lateral buckling
Hannskarl Bandel, for architect’s
insurance company – blamed faulty
weld connecting scoreboard to roof
Professional and Procedural
Aspects
Five subcontracts coordinated by
construction manager
Construction manager refused to hire a
structural engineer to inspect project
After collapse, construction manager
disclaimed all responsibility
LZA report – inspection and/or QC
inadequate, poorly handled
Ethical Aspects
Excessive deflections brought to
engineer’s attention
Engineer provided assurances, without
rechecking work
Workers knew building was a death trap
(Philadelphia Inquirer, May 28, 1978)
Higher factor of safety for highoccupancy buildings?
Legal Repercussions
Parties reached out-of-court settlement
six years after collapse
No definitive ruling on the cause of
collapse
Professional and Procedural
Aspects
Visible distortion and bowing of
structure should have been a red flag
Connecticut requires peer review of
certain buildings, but did not for this
building
Educational Aspects
Need to check whether computer
results make sense – hand calculations,
knowledge of structural behavior
Bracing and assumptions, buckling
strength
Moment of inertia
7
Conclusions
Computer results must be checked
Need field inspection
Unexpected deflections must be
explained
Avoid confusion in project responsibility
References
Martin, Rachel, and Delatte, Norbert,
“Another Look at the Hartford Civic Center
Coliseum Collapse,” ASCE Journal of
Performance of Constructed Facilities, Vol. 15,
No. 1, February 2001.
Feld and Carper, Construction Failure, Wiley,
1997
Kaminetzky, Design and Construction
Failures: Lessons from Forensic
Investigations, Mc-Graw-Hill, 1991
Levy and Salvadori, Why Buildings Fall Down:
How Structures Fail, Norton, 1992
8
In Steel Structures
most structural failures are the result of:
 Stability problems
 Connection problems
 Stability and connection problems
The Kemper Arena
Collapse
Kansas City, Missouri
June 4, 1979
1
Structure and Roof
 Opened in 1973 – new home of Kansas
City Kings basketball team
 Suspended roof provided uninterrupted
sight lines for spectators
 Roof supported by 42 hangers, made with
ASTM A490 high strength bolts
 To reduce storm water runoff into city
sewers, roof designed to hold water – only
8 5 in./13 cm diameter drains
2
Kemper Arena Collapse, June 4, 1979
Collapse
 American Institute of Architects (AIA)




awarded arena an innovative design prize
Roof collapsed at same time as AIA
opening banquet
Evening rain and wind storm – 4 ¼ in./11
cm of rain per hour, 70 mph/112 kph wind
Roof collapsed with an explosive bang
No one hurt
Photo courtesy WJE
3
Investigation – Weidlinger
Associates
 Worked on behalf of a subcontractor
 Probably many fatigue cycles over 6 years
since arena opened
 Roof susceptible to ponding, wind made
water pile up in one place
 One hanger fractured under water weight
plus wind action
 Once a single hanger failed, roof had no
redundancy
Investigation – Failure Analysis
Associates
 Worked on behalf of the steel
manufacturer
 Light roof susceptible to ponding,
vibration, other problems
 Loose roof bolts made it more flexible –
bolts were never tightened properly
 Wind drove water to weakest part of
structure
 Enough static prying force to fail bolts
4
Investigation – James L. Stratta
 Worked on behalf of owner, Kansas City
 Said bolts failed due to weakening of
ASTM A490 high-strength bolts over time
due to fatigue
Conclusions
 Light, flexible, optimized structures are at
risk of deflection and ponding
 Structures must satisfy both strength and
stiffness requirements
 High strength metals are more susceptible
to fatigue (see Point Pleasant Bridge)
Key References
 Levy, M., and Salvadori, M. (1992). Why
Buildings Fall Down: How Structures Fail,
W.W. Norton, New York.
 Wearne, Phillip (2000), Collapse: When
Buildings Fall Down, TV Books, L.L.C.
(www.tvbooks.com), New York, N.Y.
5
The
Collapse
That
Wasn’t: The
Citicorp
Tower Case
The Citicorp Tower
 Completed in 1977 at a cost of $ 175
million
 59 story, 25,000 ton steel skeleton
 Nine story high columns at midpoints of
walls, not at corners - left corner open for
a church
 Church required 100 feet of clear air
Outline
 Discovering the problem
 LeMessurier’s options
 Taking action
 Consequences
 Lessons learned
The Citicorp Tower
 Unique chevron bracing design to carry forces to
columns
 Used a tuned mass damper (TMD) - 410 ton
block of concrete on a film of oil
 TMD reduced sway in wind
 Wind tunnel tests carried out at Boundary Layer
Wind Tunnel Laboratory at the University of
Western Ontario
The Citicorp Tower
Chevron
Bracing
Discovering the Problem
 Structural engineer - William LeMessurier
 Technical article written by LeMessurier
 Chevron connections bolted, not welded
 Phone call from a student questioned design -
partner Stanley Goldstein
 Architect - Hugh Stubbins, Jr.
 LeMessurier analyzed tower under quartering
June 1978
winds for a structural engineering lecture at
Harvard
 Quartering winds would increase chevron brace
forces due to wind by 40 %
Discovering the Problem
 Problem compounded by replacement of welded
connections with bolts
 Higher forces lead to tension in chevrons - 160
% increase of net force on bolts
 Braces designed as truss members, not
columns - lower factor of safety
 Wind tunnel tests showed wind forces could
actually be higher than designed for
Wind Vibration
Discovering the Problem
 Took plans to lake home for analysis
 Key joint was at the 30th floor
 Building unsafe in 16 year wind - about 80
mph
 Building can take 52 year storm, if damper
(TMD) works
Discovering the Problem
 However, damper requires electricity - may
lose power in storm
 Now end of July, with hurricane season
approaching
 Possible litigation, bankruptcy, disgrace
Danger Area
LeMessurier’s Options
 Only LeMessurier knew extent of problem
 Suicide
 Silence - wait to get caught
 Blowing whistle on himself
Taking Action
 Contacted architect, insurance company,
met with their lawyers
 Lawyers called in another engineer, Les
Robertson to verify problem
 Robertson said TMD could not be counted
on in a storm
 LeMessurier and Stubbins contacted
Citicorp
Taking Action
Taking Action
 Citicorp chairman Walter Wriston pledged
 Supply of steel plate and availability of
full support for repairs - assigned 2 VPs
 Plan to upgrade connections by welding 2
inch thick steel plates developed
 Estimated cost $ 1 million
 Standby generators and maintenance
crew for TMD
welders critical
 Strain gages applied to critical members
 Team of weather experts assembled
 Drawings had to be approved by New York
Department of Buildings
Taking Action
Taking Action
 Publicity now became an issue - hard to explain
 September 1st, Hurricane Ella is off Cape
need to fix a new building
 Press release referred to “new results from
dynamic wind tunnel tests at UWO”
 Whole truth explained to NYC Acting Building
Commissioner
 City offered to rapidly certify new welders to
meet requirements
Hatteras, heading for New York
 Status of repairs was that building could
withstand 200 year storm, without TMD
 Hurricane moved out to sea
 On September 13th, weather watch and
evacuation plans canceled
Taking Action
 Welding finally completed in October
 Building today can withstand 700 year
storm, without TMD
Consequences
 Citicorp served notice on LeMessurier and
Stubbins
 LeMessurier estimated cost $ 4.3 million
 HRH Construction estimated $ 8 million
 No litigation - Stubbins was held harmless,
LeMessurier gave up $ 2 million insurance
 LeMessurier’s insurers later lowered his
premium
Lessons Learned
Lessons Learned
 Engineers make mistakes - the true test of
 Exceptional projects require exceptional
character is how you handle them - Canon 6
 Public safety must always be paramount Canon 1
 Meeting the building code does not ensure you
have a safe structure - building codes are for
ordinary structures and ordinary uses - Canon 1
engineering efforts - Canons 1 and 2
 Keep the client informed, particularly of
bad news - Canons 4 and 6
 Fix the problem first, then fix blame Canon 1
 Continuing education is critical - Canon 7
Lessons Learned
 Information about failures as well as
success stories must be passed on to the
profession - Canon 7
References
 Morganstern, J., “The Fifty-Nine-Story Crisis,”
ASCE Journal of Professional Issues in
Engineering Education and Practice, Vol. 123,
No. 1, Jan. 1997, also in the May 29, 1995 issue
of The New Yorker
 ASCE Code of Ethics,
http://www.asce.org/aboutasce/ethics.html
References
 The Online Ethics Center for Engineering
& Science, onlineethics.org
 LeMessurier, W., Course notes for MIT
1.544 Structural Design of Buildings, Fall
1985
Web References
 http://www.asce.org/inside/codeofethics.cf
m
 http://onlineethics.org/moral/LeMessurier/l
em.html
 http://www.lemessurier.com/
 http://www.cet.nau.edu/Academic/Design/
D4P/EGR286/Course_Materials/Homewor
k/ethics.htm
Sampoong Super Store
Collapse
Design and Construction
 Opened in December 1989
 Five stories above grade, four stories
below grade
Seoul, South
Korea
June 29, 1995
Major Changes
 Built on a landfill – concerns about
foundation
North Wing Plan– from Gardner et al.
 Sampoong erected building after Woosung
Construction completed foundation and
basement
 Fifth floor added
 Converted from office block to department
store – higher loads
 Changed upper floor from skating rink to
traditional Korean restaurant with floor
heating
Warnings of Collapse
 Store passed safety inspection in June
1995
 Shortly after, cracks in walls and water
pouring through holes in ceiling
 June 29, structural engineers inspected
and declared building unsafe
 Company executives overruled engineers
– patched cracks, moved heavy goods to
basement
Collapse
 Store collapsed in about 10 seconds at 6
pm on June 29
 Only façade left standing
 Overall death toll 498
 One survivor found alive 17 days after
collapse
1
From www.skyscrapercity.com
From http://channel.nationalgeographic.com/series/seconds-from-disaster/2719/Overview
Collapsed Store – from Gardner et al.
Investigation
 Professors Chung and Choi, Dan Kook
University and Soongsil University
 Flat slab structure without cross beams –
inherently nonredundant
 Adequate foundation
 Large chunks of structure cut away in
conversion from office block
 Fifth floor restaurant added considerable
weight – 3 ft./90 cm thick concrete
2
Investigation
Design/Construction Errors
 Restaurant floor plan not compatible with
 Columns only 24 in./62 cm vs. 35 in./89
lower floors – columns did not line up
 Large, heavy water cooling blocks on roof
 To support water coolers, slab was
thickened – but no columns were added
 Recently water cooling blocks had been
moved across roof, causing cracks up to 1
in./ 25 mm
cm, only 8 of 16 reinforcing bars provided
 Slab dead loads miscalculated – based on
4 in./10 cm but actual slabs 3 or 4 times as
thick
 Some reinforcement left out, poor
slab/floor connections
 Spans between columns increased to
almost 36 ft./11 m for more sales space
Investigation Conclusions
Punching Shear Capacity –
Gardiner et al.
 Prof. Chung and colleagues
 Human ignorance, negligence, and greed
 Illegal alteration of architectural design
and use of building
 Poor supervision by planning authorities
 Management did not act on structural
problems observed over 5 years of use
Legal Actions
 Joon Lee, chairman of Sampoong, and
son Han-Sang Lee, sent to prison for 10 ½
and 7 year terms
 Twelve building officials convicted of
taking bribes, up to $ 17,000 US
 Investigations found widespread patterns
of corruption in Korean building industry
 Reduction in concrete strength
 Reduction in effective slab thickness
 Reduction in column diameter
 Omission of a drop panel, reducing slab
thickness from 18 to 12 in./45 to 30 cm
Key References
 N.J. Gardner Jungsuck Huh, Lan Chung,
(2002), “Lessons from the Sampoong
department store collapse,” Cement &
Concrete Composites volume 24 (2002)
pp. 523–529, Elsevier
 Wearne, Phillip (2000), Collapse: When
Buildings Fall Down, TV Books, L.L.C.
(www.tvbooks.com), New York, N.Y.
3
MIT
Hotel Vendome Fire
Collapse – Boston June
17, 1972
 Course 1.413 Construction Technology
 Master’s students in civil engineering,
architecture, real estate development
 Case studies
 Collapse of Hotel Vendome during a fire
 Collapse of 2000 Commonwealth Avenue
during construction
 Course 1.544 Structural Design of
Buildings with Bill LeMessurier
Hotel Vendome
Hotel Vendome as Built, 1880
(Wikipedia)
 Built in 1872 at 160 Commonwealth




Avenue, Boston
Five story building, 42 by 103 ft.
Modeled on Hotel Vendome in Paris
Masonry bearing wall structure with
wooden floor joist construction
Founded on wooden piles with granite
caps
History
Hotel Vendome circa 1910
 New wing, seven stories and tower, 1881
 Exterior façade cut stone to 6th floor
 Steeply sloped Mansard roof finished with
slate
 Major alterations in 1890
 Two masonry walls removed on first floor
 Replaced with wrought iron I beams and cast
iron columns
1
History
Renovation Plans
 “Solarium” added over northeast corner of
roof in 1911
 Prestigious hotel from 1911 to 1960
 First public building in Boston to have electric
lights
 Notable guests General US Grant, President
Grover Cleveland, Mark Twain, Oscar Wilde,
Sarah Bernhardt
 Series of six fires in 1968 and 1969
 Occupancy certificate revoked
 Investors decided to renovate building,
1970
 Renovation and reopening of Café
Vendome
 2nd phase convert test of building to luxury
apartments and shopping mall
 Modifications included installation of
ventilation ducts below first floor structure
Boston Building Department
From MIT course
case study notes
 April 28, 1971 – “short form” permit issued
to stripping and rubbish removal prior to
Café Vendome renovation
 May 24, 1971 – application 664 “long
form” permit submitted for renovations
 June 30, 1971 – application 664
abandoned, replaced by application 860,
changed occupancy to “hotel”
Boston Building Department
New Boston Building Code
 Application 860 submitted with note “no
 In effect July 1, 1971
 New code would have increased delay
structural changes… for the architect”
made by person not an architect or
engineer
 Permit issued without formal plan review
 Also submitted application 862 for
renovation into apartments and shopping
mall, processing delayed for zoning
and cost
 Pressure to obtain permits before new
code went into effect
2
Renovations
 Work began July 15, 1971 by HVAC
subcontractors
 Application 860 approved July 26, 1971,
permit issued
 Investors told architect and engineer for no
further work on August 3, 1971, costs
exceed budget
 Mid-August – HVAC openings made in
basement, no structural engineer
consulted
Renovations
 Boston Building Department inspected
August 15, 1971, found work exceed
permit
 Violation notice prepared and issued
 September 15, 1971, architect and
engineer brought back on
 October 7, 1971, project manager saw
HVAC openings in basement,
recommended repair
Renovations
 Boston Board of Appeals allowed variance
to zoning code, October 26, 1971,
preservation of architecture and hotel use
 Permit for application 862 granted
December 6, 1971
 Café Vendome opened December 7, 1971
 Second violation issued December 15,
1971, no licensed builder on premises
Inspections
 25 inspections up to June 13, 1972
 Probably visits no more than 45 minutes,
short entries in log
 Inspector did not refer to the plans
 No structural deficiencies observed
Fire and Collapse
Fire and Collapse
 Fire found at 2 pm on 4th floor on June 17,
 Fire reported under control 5:20 pm
 Much of structure was burned out
 Collapse at 5:30 pm without warning at the
1972
 By 4 pm, 26 firefighting companies
responded
 Estimated 135,000 gallons (over 1 million
pounds) of water pumped onto building
 Ponding observed on upper floors as they
sagged
southeast corner
 Collapse section had less fire damage
 Area of internal iron column
 Nine firemen killed and eight injured
 Last body recovered the next day, Father’s
Day
3
The Collapse – from
www.fireengineering.com
http://www.fireengineering.com/articles/
2012/11/10-15-tragedy-on-com-ave-thecollapse-of-the-hotel-vendome.html
From MIT course
case study notes
Firefighter Memorial – from
CelebrateBoston.com
Hotel Vendome Fire Collapse
 The… Memorial recalls the tragedy that occurred just a few steps
away on Father's Day, June 17, 1972. The fire at the Vendome
Hotel claimed the lives of nine firefighters and injured many more.
This fire remains the largest loss-of-life incident in the history of the
Boston Fire Department.
(http://www.bostonfirehistory.org/firefightermemorials.html)
Investigation
 Duct opening found in bearing wall just
under iron column
 Bearing wall collapsed under column
 Stress of iron column on bearing wall was
4 to 5 times allowable









Fire Fighter Thomas W. Beckwith
Fire Fighter Joseph F. Boucher
Lieutenant Thomas J. Carroll
Fire Fighter Charles E. Dolan
Lieutenant John E. Hanbury, Jr.
Fire Fighter John E. Jameson
Fire Fighter Richard B. Magee
Fire Fighter Paul J. Murphy
Fire Fighter Joseph P. Saniuk
Source Material
 Seymour, W., Irwig, H.G., and Becker,
J.M. (1984) “A Collapse at the Hotel
Vendome – Case Study of a Failure in
Building” MIT
4