<< Back to Construction Incidents Investigation Engineering Reports

For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Investigation of the August 5, 2002, Collapse of
Tilt-Up Precast Concrete Wall Panel in
Greensboro, NC


U.S. Department of Labor
Occupational Safety and Health Administration
Directorate of Construction

September 2002

Investigation Reports, Investigation of the September 10, 2012 Partial Collapse of a Building Under Construction at 227 Carlton Avenue in Brooklyn, NY

Investigation of the August 5, 2002, Collapse of
Tilt-Up Precast Concrete Wall Panel in
Greensboro, NC

September 2002

Report prepared by
Mohammad Ayub, PE
Dinesh Shah, PE

Table of Contents

  1. Introduction
  2. Field observation and measurements:
  3. Stability of the Wall
  4. Conclusion
  5. Figures
Report

Introduction:

The incident occurred on Monday, August 5, 2002 at about 11:30 AM in Greensboro, NC when a tilt-up concrete wall, weighing approximately 40,000 pounds, suddenly fell over three workers eating their lunch in the shade of the wall on a hot day. Approximately two hours earlier, the tilt- up wall sub-contractor, at the direction of the tilt-up wall contractor, had removed the pipe braces of the tilt-up wall, based on the report from the Testing Agency that the welds between the roof joists and the embed steel plates of the wall have been completed. As the welds were reportedly completed, the tilt-up contractor determined the braces to be redundant and hence removed them, though the bottom of the wall was not structurally supported (figure 3). There were shims placed under the wall at each end of the panel. As will be discussed later in this report, the Testing Agency's report was flawed because, in fact, the welds were not properly done and were ineffective in providing any lateral support to the wall at the top. There were no lateral supports of the wall at the bottom except for the friction between the concrete and the shims. At the bottom, the wall was neither tied to the slab on grade nor grouted over the foundation wall.

The failed wall was part of the perimeter of the building for Home Depot Store (figures 1 & 2), under construction at 2912 South Elm-Eugene Street, Greensboro, NC. The structure is comprised of steel framing with steel columns, steel joist girders and joists. The joists of the exterior bays were supported over the tilt-up wall panels. The panel in question was 6 1/2" thick, approximately 23 feet high and 20 feet wide. It was supported over the foundation wall over shims at each end. The adjoining panels on either side of the failed panel were of similar construction and were also supported on shims. Later, the space between the tilt-up wall and the foundation wall was to be grouted with high strength grout; and the tilt-up wall panel was to be connected to the slab on grade by a closure strip i.e. comprising of the last 4' of the slab on grade (figure 3). At the top of the wall, there were three steel plates embedded in the wall to support the steel joists. The embedded steel plates were 6 1/2" x 12" x 1/4".

Field observation and measurements:

Wall:

The wall fell on the west side. It was saw cut in two pieces of 18'-1" and 4'-11" during the rescue phase. The wall measured 19'-7 5/8" in width and 23' in height. The thickness was measured to be 6 1/2". See attached field measurements and shop drawing of the panel (figures 3, 4, & 5). No discrepancy was noted in the fabrication of the panel. The embed plates were measured to be 6 1/2" x 12" x 1/4", as called out on the contract drawings.

Embedded plates:

The three embed plates (figure 4, section E) were examined to determine whether or not any welding was done. The north embed plate shows a 1/2" long tack weld on the north side and a 1 1/2" long weld in the north-south direction. Please note that the welds were to be made in the east-west direction along the seat of the joist. The middle embed plate did not exhibit any weld.

The south embed plate showed a weld in the east-west direction but only on one side, i.e., on the south side. This weld did not indicate any failure. In addition, there was a deposit of weld at a comet of the plate, which was away from the bearing of the joist (figure 4, section E).

The three corresponding joist bearings were then examined. The joist bearing on the north end of the wall did not exhibit any significant remnant of the weld. It is therefore concluded that the tack weld seen on the north embed plate had no structural significance. The middle joist bearing angles were void of any welds. This observation corresponded with the findings of the middle embed plate. The south joist indicated that the north end of the bearing angles was welded. The south end of the bearing was observed to be clean. It is interesting to note that the south embed plate indicated welds on the south side where as the joist bearing angles showed welds on the north side. It is believed that the welder had at one time welded the joist at a location corresponding to the welds on the embed plate but later "burned" them. It appears that the joist bearing was later moved a few inches, and the welding at the new location was not completed.

Interestingly, it was reported that immediately after the failure, the welders were noticed welding the joists on the panels on either side of the failed panel, while rescue was still going on. It is strongly suspected that the adjoining panels were also not welded.

Shims:

The number of shims placed under the failed wall could not be ascertained as they were scattered around. However, the shims placed under the panel north to the failed panel indicated that there were only three shims placed under the wall. Two were placed on the north end and only one was placed on the south end.

Grout:

The structural drawings called for 1" grout between the underside of the tilt-up wall and the top of the foundation wall (figure 2, section B). However, field observation indicated that the space for the grout varied from 2" to 4". This indicates that the construction of the concrete foundation wall was not done as per the standard practice in accordance with the contract drawings. In addition, the contractor did not grout the space beneath the panel soon after the erection of panel wall as recommended by the Tilt-Up Design and Construction Manual of Tilt-Up Concrete Association (attachment C, page 2).

Stability of the Wall:

The wall derives its stability from the roof and floor diaphragm, after the construction is completed. Until such time, temporary braces are provided to prevent collapse and failure. In this case, the tilt-up wall contractor was provided with the information (attachment A) from the Testing Agency that led him to believe that the top permanent connections have been made. The bottom connection was, however, not completed as the contractor had yet to finish casting the last strip of the slab on grade, which was to tie the wall to the horizontal diaphragm. Besides, the grout underneath the wall was not placed. The only support available to the wall at the bottom was the shims without any positive connection either to the wall or to the foundation wall. The friction between the shims and the concrete wall cannot be relied upon to provide stability to the wall, as there were no positive connection between the shims and the wall, and shims were subject to movements. Field observations of the two impressions left on the underside of the failed tilt-up wall panel indicated that shims were not placed directly under the wall but under the cantilevered heel of the panel. Therefore, when the temporary braces were removed, the shims did not provide even a minimal degree of stability to the wall, and failure occurred soon thereafter.

The contract specification on erection, LA-121-5619-00, Part 3, Section 3.06, Paragraph G, Note 5 (attachment D) states "provide adequate bracing until all structural steel, purlins, beams, joist and deck is placed, backfill placed and floor slab leave out placed."

North Carolina OSH standard 1926.704 (a) states "Precast concrete wall units, structural framing, and tilt-up wall panels shall be adequately supported to prevent overturning and to prevent collapse until permanent connections are completed."

American National Standard, ANSI A10.9-83, Section 9.4 states "Precast concrete wall units, structural framing, or tilt-up wall panels shall be braced until permanent connections are completed."

The Tilt-Up Design and Construction Manual of Tilt-up Concrete Association (attachment C, page 4) states that the temporary braces should be removed after the roof structure and floors have been connected to the wall so that the permanent lateral bracing system is operational.

The General Notes of contract drawing S1.0 on Tilt-Up panels, note 10 (attachment B) states "...The contractor is solely responsible for providing all necessary bracing as required for construction loads, for stability, and for resistance to wind and seismic forces until the entire structure is complete...."

Conclusion:

  1. The failure occurred because the contractor prematurely removed the temporary braces of the tilt-up wall panel before permanent connections at the top and bottom of the panel were made. This was a violation of the contract specification, North Carolina OSH Standard, ANSI Standard, and tilt-up wall industry standards.

    The stability of the wall is derived from the connection between the wall and the roof joists, and between the wall and slab on grade or foundation wall. The temporary bracing must only be removed after all permanent connections are completed i.e. grout placed under the panel, closure strip of the slab on grade cast, and the connections at the top of wall completed.
  2. The Testing Agency erred in providing a report that the welded connections between the wall and the joists have been completed. This is not supported by the facts discovered after the failure.
  3. The tilt-up contractor did not construct the foundation wall as per the construction documents. Instead of 1", the space of the grout was observed to be to 2 to 4". Further, the location of the shims under the wall did not follow any uniform pattern. In addition, the contractor did not grout the space beneath the panel soon after the erection of panel wall as recommended by the Tilt-Up Design and Construction Manual of Tilt-up Concrete Association.

Figures:

Figure 1 - Schematic Plan (Boundary of Tilt-Up Concrete Wall Plan) - For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Figure 1 - Schematic Plan (Boundary of Tilt-Up Concrete Wall Plan)


Figure 2 - Tilt-Up Wall Panel Elevation A-A, Section B (At Cottom of Tilt-Up Panel), Section C (At Top of Tilt-Up Panel) - For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Figure 2 - Tilt-Up Wall Panel Elevation A-A, Section B (At Bottom of Tilt-Up Panel), Section C (At Top of Tilt-Up Panel)


Figure 3 - As Built Data Measured in the Field - For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Figure 3 - As Built Data Measured in the Field


Figure 4 - Section D (Shim Plate Location at Bottom of Tilt-Up Panel), Section E (1/4" Thk Embed Plate Location at Top of Tilt-Up Panel) - For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Figure 4 - Section D (Shim Plate Location at Bottom of Tilt-Up Panel), Section E (1/4" Thk Embed Plate Location at Top of Tilt-Up Panel)


Figure 5 - For problems with accessibility in using figures and illustrations, please contact the DOC at 202-693-2020.

Figure 5


Attachment A (for the complete attachment, see PDF)

Attachment B (for the complete attachment, see PDF)

Attachment C (for the complete attachment, see PDF)

Attachment D (for the complete attachment, see PDF)


Accessibility Assistance: Contact the OSHA Directorate of Construction at (202) 693-2020 for assistance accessing PDF materials.