L'Ambiance+Plaza

L'Ambiance Plaza (April 23, 1987) toc

// Liam J. McNamara, BAE/MAE, Penn State, 2009 // //Expanded by M. Kevin Parfitt, P.E. Penn State, July 2010// =**__ Introduction __**=

On April 23, 1987 the L’Ambiance Plaza building, in Bridgeport, Connecticut, collapsed during construction. This collapse spurred a large scale, eight day, rescue attempt and ultimately left 28 workers dead.(Moncarz, 1992) This 16 story building, 13 apartment levels over 3 parking levels, was being constructed using the lift-slab method. The lift-slab method consists of casting post-tensioned floor slabs, one on-top of another, at ground level and then hydraulically jacking each level into place. At approximately 1:30pm on April 23rd a loud bang was heard and within the next 2-10 seconds the entire building crashed to the ground. The collapse launched several investigations but was quickly settled out of court ending all investigations and leaving the exact cause of the collapse unknown. Although the exact cause of the collapse remains unknown, five viable theories have been proposed in the years since the collapse by various experts on building failures.(Schribner, 1988)



Description of Building
L'Ambiance Plaza was planned to have two virtually identical towers, with floor plans measuring approximately 63 ft by 112ft, and a neighboring parking garage. The parking garage had not begun construction and was not involved in the collapse. (Korman, 1988) The towers were separated by 4 ft and would have been joined by cast-in-place concrete during the final stage of construction. The structural system consisted of steel columns and 7" thick two-way unbonded post-tensioned concrete flat plates with shearwalls at four perimeter adn four interior locations (Culver, 1987). Generally the Youtz-Slick method of lift-slab construction, which was employed on L'Ambiance Plaza, is used in the design of two to five story buildings. However, there have been taller structures built using this method.(Cuoco, 1992) The towers are designated the east and west buildings and will be referred to as such throughout this case study.

The Lift-Slab process used at L'Ambiance Plaza consisted of casting floor slabs at ground level, raising those floor slabs to the desired elevation with hydraulic jacks, and fixing the slabs in position mechanically. By casting the floor slabs on grade, the need for shoring or formwork underneath each slab was eliminated, and only side forms were required. Slabs were raised in packages and parked at designated levels by a prescribed schedule of construction. The packages were "parked" at a level by 3 workers, one operating the jack, one monitoring the level from underneath, and a third placing in steel wedges and tack welding them in place. Packages of floor slabs were "parked" at various levels until shearwalls could be cast to provide stability, and steel erection could continue to higher levels. (Culver, 1987).

Day of the Collapse
The post-tensioned floor slabs were grouped in packages of 2-3 floors each, which were to be jacked to various levels during the construction period before being set at their final assigned level. These positions were preplanned in accordance with stages of construction. The morning of the collapse the building was in Stage IV of construction. At this point in Stage IV, the west building package containing floor slabs 9-11 were jacked into position at level six approximately 8" below the package with the slabs for floor 12 and the roof(Cuoco, 1992). The ironworkers were working to tack weld wedges under the 9-11 package to temporarily hold it in place. They were working on this through out the day. After lunch at approximately 1:00pm workers used a 12-ton horizontal jack between the two towers to plumb the West building(Masih,1995). Around 1:30 pm the ironworker, Kenneth Shepard (Martin, Delatte, 2000), who was installing wedges in the west building, heard the first loud bang and looked up to see the concrete above him "cracking like ice breaking." (Cuoco, 1992) The floor slab above him the proceeded to collapse onto the levels below. Within 2-10 seconds both towers had collapsed completely. The west building fell first, followed shortly by the east building. Both buildings collapsed in a similar, "pancake" manner settling almost entirely with in the plan of the building towards their respective centers.

=__Causes of Collapse__=

Theory 1: Instability of the wedges supporting the 12th floor and roof package
-Thornton Tomasetti Engineers

Thornton-Tomasetti Engineers' concluded the epicenter of the collapse was a core column, 3E, of the west building. Wedges supporting the 12th floor and roof package at the column were unstable and started the collapse. They state in their "Collapse Scenario" that a wedge supporting the 12/R package rolled out leaving the shearhead at this level supported by a single wedge. The horizontal load from the jack used to plumb the building caused the remaining wedge to roll slightly as evidenced by rounding and bending in the west weld block of the shearhead. Additional movement of the slabs may have caused the remaining wedge to roll completely out. (Cuoco, 1992)



In the absence of both wedges the package would have dropped the 8 in. onto the top of the package containing floors 9-11. The loud bang was heard when the lifting nuts initially supporting only floors 9-11, slipped out under the additional load and impacted the web of column 3E. The slabs and shearheads then began to slide down the column, impacting the floor slabs below, ripping them from the column, and progressively collapsing the west tower. The catenary action of the post-tensioned cables then pulled the remaining columns towards column 3E. The east building then collapsed from horizontal forces transmitted through the pour strips or horizontal jack, or impact of the west tower debris. (Cuoco, 1992)

The physical evidence supporting their theory included the discovery of abnormal tack welds on the wedges supporting the 12/roof package as well as shearhead gaps on columns 3E and 3.8E(0.628 in.) that were much larger than those on the rest of the building (0.233 - 0.327 in.) and other buildings built using the lift-slab method (0.250-0.375in). These large gaps as well as the presence of hydraulic fuel would have reduced the friction normally depended upon to hold the wedges in place until they can be completely welded.(Cuoco, 1992)

Theory 2: Jack rod and lifting nut slipped out due to a deformation of an overloaded steel angle welded to a shear head arm channel
-National Bureau of Standards (NBS)

The NBS concluded in their investigation that the failure began at the building's most heavily loaded column, E4.8 or the adjacent column, E3.8, as a result of a lifting assembly failure. At each column the shearhead reinforces the concrete slab, transfers vertical load from the slab into the column, and provides a place of attachment for the lifting assembly. Steel channels are cast in the slab, allowing room for the lifting angle. Lifting rods, raised by hydraulic jacks above them, are passed through holes in the lifting angle and fastened with lifting nuts. (Scribner, 1988)



NBS testing determined that when the shearhead and lifting angles were loaded with forces nearing 80 tons, they had a tendancy to twist. This was due to a lack of stiffness, not strength. During the lifting process the shearheads and lifting angles were loaded close to their maximum capacity. The angles deformed under the excess force of the three 320 ton slabs, causing the jack rod and lifting nut to slip out of the angle and hit the column. This contact would have produced the loud bang heard by Kenneth Shepard. (Martin, Delatte, 2000)

The building then collapsed in the same method as described by T-T after the lifting nuts impacted the column.

Theory 3: Improper design of post-tensioning tendons
-Schupack Suarez Engineers, Inc.

Schupack Suarez Engineers examined the unusual layout of the post-tensioning tendons in the west building. The east building's tendons were run in a typical two-way banded layout, uniform tendons running North - South carry the slab load to the East- West Column line, where the E-W banded tendons then "pick up" the load and transfer it to the columns. However, at column 4.8E in the west building, the E-W tendons split around the column line. The absence of tendons in Line E due to the split at the column added increased load into the structure. The design details also did not include the location of the shear walls or the openings for the walls at Columns 11A, 8A, and 2H(Poston, 1991).

Finite-element analysis determined that the tensile stresses along Column Line E, east of Column 4.8E, exceeded the cracking strength of the concrete. By this reasoning once a crack was initiated it would spread immediately to Column 4.8E. The finite-element analysis also showed that even under ideal lifting circumstances column 2H would have had unsuitably high compressive and punching shear stresses.(Poston, 1991)

Theory 4: Substandard welds and questionable weld details
-Occupational Safety and Health Administration (OSHA)

OSHA stated that two weld design details were questionable. These welds were the 1/2 in. single-bevel-groove weld between the arm channel and the lifting angle in the shearheads, and the one-sided square-groove weld connecting the header bar and header channel at the shearhead. Both welds were of unknown and unspecified penetration depth and therefore had an unpredictable strength. The single-bevel-groove weld could also have been further weakened from flush grinding. The one-sided square-groove welds are even more questionable due to the fact that they were not among the American Welding Society prequalified joints.(McGuire, 1992)

OSHA investigators found a shearhead with a failed single-bevel-groove separated completely from the arm channel. This shearhead belonged to column E3.8 previously identified as a possible epicenter for the collapse. Upon further investigation by OSHA hired consultant firm, Neal S. Moreton and Associates, it was determined that of 30 welds investigated at Column E3.8, levels 7,8, and 10; 17 welds did not meet industry standards.(McGuire, 1992)

Theory 5: Global instability caused by lateral displacement
-Failure Analysis Associates, Inc. (FaAA)

FaAA consultants focused on the response of lateral loading and overall torsional instability. The shearhead connection is rotationally stiff when the concrete slab is temporarily resting on the wedges and when it is fully welded in its final position it becomes a rigid connection. However, when the slab is lifted off the wedge it can rotate freely. In the absence of lateral loading the building would be completely stable. In the presence of lateral loading or displacement, such as that from the horizontal jacking just after lunch, the slab could be lifted off a wedge and the building would become laterally flexible. FaAA used 3D computer modeling (ANSYS) and nonlinear stability modeling to investigate this possibility. Upon analysis of their modeling FaAA concluded that lateral instability was the cause of collapse for both the west and east buildings.(Moncarz, 1992)

=__Conclusion__= = = Due to the relatively quick settlement of the case, many possible lessons will never be learned from this terrible collapse. The lift-slab method whose use is responsible for approximately 45,000,000 sqft of safe buildings, suffered greatly from this collapse, never making a come-back in the United States (Cuoco, 1992). The collapse calls into question why critical connection details don't get enough attention from engineers, as well as revealing a need for greater attention to temporary construction load. This failure shows the need for clear connection details during the design phase, as well as a need for lateral bracing during construction, and strict following of the specified construction phases (Korman,1987). Collapses such as L'Ambiance Plaza should be used as building tools for future construction, the building industry must continue to learn from its mistakes so as to prevent them in the future.

A lack of communication between several subcontractors and the engineer was also a major issue in the L'Ambiance Plaza collapse. Responsibility for design was fragmented among so many subcontractors that several design deficiences went undetected. If the engineer of record had taken responsibility for the overall design of the building or a second engineer had reviewed the design plans, these defects probably would have been detected (Heger 1991). The L'Ambiance Plaza case was mediated by a two-judge panel who determined a univeral settlement between 100 parties, which closed the case. Twenty of more separate parties were found guilty of "widespread negligence, carelessness, sloppy practices, and complacency." All parties contributed to the $41,000,000 settlement fund in various amounts. The families of those killed in the collapse, and the workers injured received $30,000,000 in the settlement (Martin, Delatte, 2000.)

Prior to the collapse of L'Ambiance Plaza, Connecticut had no provisions in their building regulations for an independent review of building structural designs. Generally, building authorities do not have an adequate staff to review structural designs, and there were no requirements for peer review. An improvement made in Connecticut as a result of the collapse was to require review of structural designs by an independent engineer accepted by the building authority (Heger, 1991).

=Other Lift Slab Structures of Note:=

Although the collapse of L'Ambiance Plaza effectively ended the use of lift slab construction in the US, a number of structures had been built prior to April 23, 1987. Many of these were successful and are still in service today. Others experienced problems such as column-wedge failures, global instability (sidesway failure mode) or related construction issues that resulted in collapse or remediation during construction. Fortunately there was no loss of life with these cases, however there were injuries and ssignificant economic losses. Examples of several problem Lift-Slab are noted below:
 * On July 15, 1954, the Junipero Serra High School Roof in San Mateo, California collapsed due to sidesway instability. This 16 ft. tall one-story building was approximately 65 ft. x 70 ft. in plan and used 6 inch diameter steel pipe columns for support. Attempts by the contractor to stablilze the leaning structure with guy wires on one side of the frame backfired as the contractor apatently overcompensated and failed the building in sidesway in the direction opposite the original lean(Zallen and Peraza 2003, pg 24)


 * A structure based on the Canadian wedge system (a system that relies on frictional resistance between the column and wedges) collapsed under construction in Marion, Indiana in 1962 (Delatte 2009 pp 120-121).


 * Cleveland, Ohio experienced a near collapse on April 6, 1956 when the Pigeonhole Parking Garage (using the Youtz-Slick lifting system) nearly collapsed in winds of 35-65 miles per hour prior to the steel wedges being permanently welded. Fortunately the contractor was able to right the structure by pulling it back into plumb and finish the project without further incident even though the the building was leaning as far as 7 feet out of plumb (Zallen and Peraza 2003, pg 24-25; Delatte 2009 pg 121). An expanded account of the [|Pigeonhole Parking Garage] case, including dramatic original photographs of the leaning structure, can be found on the [|Failures Case Studies website] of the (NSF MatDL) created by Dr. Norbert Dellatte.

=Early Lift Slab Construction:=

Photographs of what is believed to be one of the earliest, if not the first, lift slab structure constructed in the US, are shown below. Provided by former Professor Vincent L. Pass, P.E. of Penn State AE, the photographs are that of the Trinity University Library, San Antonio, Texas and were taken in April of 1951.



= __Bibliography:__ =

Cuoco, D., Peraza, D., Scarangello, T., “Investigation of L’Ambiance Plaza Building Collapse.” //Journal of Performance of Constructed Facilities//, November, 1992.(Pages 211-230)

Moncarz, P., Hooley, R., Osteraas, J., Lahnert, B., “Analysis of Stability of L’Ambiance Plaza Lift-Slab Towers.” //Journal of Performance of Constructed Facilities//, November, 1992.(Pages 232-245)

McGuire, W., “Comments on L’Ambiance Plaza Lifting Collar/ Shearheads.” //Journal of Performance of Constructed Facilities//, May, 1992. Sub-reference: Erratum: McGuire, W., “Comments on L’Ambiance Plaza Lifting Collar/ Shearheads.” //Journal of Performance of Constructed Facilities//, (May, 1992.), August, 1992. (Pages 78-94)

Poston, R. W., Feldmann, G.C., Suarez, M.G., “Evaluation of L’Ambiance Plaza Posttensioned Floor Slabs.” //Journal of Performance of Constructed Facilities,// May 1991. (Pages 75-91)

Masih, R., “Dynamic Force Effect on Collapse of L’ambiance Plaza.” //Journal of Performance of Constructed Facilities,// May 1995. (Pages 129-135)

Martin, R., Delatte, N.J., “Another Look at the L’Ambiance Plaza.” //Journal of Performance of Constructed Facilities,// November 2000.(Pages 160-165)

Korman, R., “Mediated settlement seeks to close the book on L’Ambiance Plaza.” //Engineering News-Record,// November 1988.

Korman, R., “Flawed connection detail triggered fatal L’Ambiance Plaza collapse.” //Engineering News-Record,// October 1987.

Culver, C.G., “Investigation of L’Ambiance building collapse in Bridgeport, Connecticut.” //U.S. Department of Commerce,// 1987. (Pages 1-37)

Schribner, C.F., "Investigation of the Collapse of L'Ambiance Plaza." //Journal of Performance of Constructed Facilities,// May 1988. (Pages 58-79)

Heger, F.J., "Public-Safety Issues in Collapse of L'Ambiance Plaza." //Journal of Performance of Constructed Facilities//, May 1991. (Page 106)

Dusenberry, D "L'Ambiance Plaza - Case Study"-NIST []