The Collapse
"Beams that were 10 m (32 ft) long and weighing between 2.3 and 4 Mg (2.5 and 4.5 tons) were being erected at the time of the collapse. A temporary connection failed, causing a chain reaction that pulled down some 60 to 70 steel members, metal decking, and sheeting covering an area 20 x 20 m (70 x 70 ft). More than a dozen workers were on the beams at the time; some fell or rode the beams over 15 m (50 ft) to the ground." (Construction failure By Jacob Feld, Kenneth L. Carper)
Below is a newspaper article with the breaking news of the collapse:
DAILY NEWS-RECORD, Harrisonburg, Va., Thursday, November 4, 1993, Page 4.Worker Killed In Building Collapse
CHICAGO--Two floors being built in the city's new main post office collapsed in a tangle of metal girders Wednesday, killing one worker and injuring six others.
James Cooper, who was working on the building near the section that collapsed, said, "It was just a normal day. I heard a crack, and it seemed the world was coming down."
Fire Commissioner Raymond Orozco would not speculate on what caused the collapse. Neither would a spokeswoman for Clark Construction Group Inc. of Bethesda, Md., parent of one of the two construction companies working on the building.
Construction on the section that collapsed had progressed to three floors and only had the steel framework and corrugated metal in place. Orozco said some workers fell from the second and third floors.
Three men were in critical condition at Cook County Hospital. Two other men were in serious condition and another was in fair condition.
What Went Wrong
From matdl.org:
"The Architect/Engineer of the project designed the steel structure of the building as well as its connections. When the steel fabricator/erector reviewed the contract drawings and specifications, he decided to change the construction sequence of some of the beam to column connections to simplify the construction work. This seemed reasonable since the structural design of the connections would not be altered.
The beam-to-column connection construction sequence as redesigned by the steel erector consisted of connecting two plates to the web of the beam using 30, 1-in.-diameter bolts in the shop, then connect a temporary erection angle to the web of the column with two, 1-in.-diameter bolts with nuts in the shop. Then, in the field, the bottom flange of the beam was to be placed on the temporary erection angle at a distance of 1-in.-diameter from the face of the column web and connected the beam flange to the temporary erection angle with two, 1-in.bolts with nuts. At the end, to finish the connection, the two connection plates (previously bolted to the web of the beam) were to be welded to the web of the column and the temporary erection angle disconnected. This connection construction sequence is typical for low-rise structural steel frames.
After an in depth investigation it was revealed that important changes were made to the temporary beam to column connection construction. First, the beams were placed 1 1/4 in. from the face of the columns instead of the specified 1 in. on the erector design. The 1/4 in. difference increased the eccentricity of the load from the beam on the erection angle and prevented the use of the specified two 1-in.-diameter bolts with nuts to connect the beam to the temporary erection angle. Only one 3/4-in.-diameter bolt was used to connect the inferior flange of the beam to the temporary erection angle. The 3/4-in.-diameter bolt was installed without the corresponding nut; this considerably reduced the safety of the connection. On-site investigation revealed that on 44 of 47 similar connections only one 3/4-in.-diameter bolt without nut was placed.
Finite element analyses showed that if the 3/4-in.-diameter bolt would have been used with a nut the ultimate capacity of the erection angle would have been increased and the failure might have been prevented.
The steel fabricator/erector of the project was fined by the Occupational Safety and Health Administration, charged with two misdemeanors by the U.S. Department of Justice and faced criminal charges leveled by the U.S. Department of Labor. Claims by the families of the killed and injured workers in the collapse were settled."
Consequences
The Architect/Engineer of the project was cleared of any liability for the accident.
According to the Aurora OSHA Construction News:
"On February 18, the U.S. Court of Appeals for the Seventh Circuit (Chicago - No 98-1767)) affirmed that Pitt-Des Moines (PDM) was willfully in violation of the OSHA Act. The ruling was in the case of the 1993 steel structure collapse of the post office project in downtown Chicago. The conviction and $1 million penalty were affirmed. In addition, the issue of multi-employer was raised for the non-PDM employee. The Appeals Court ruled: "The doctrine holds that on multi-employer work sites, an employer who creates a safety hazard can be liable under the Act regardless whether the employees threatened are its own or those of an another employer on the site."(page 6) "Without the doctrine, employers could avoid OSHA liability for the hazardous conditions they create merely because the threatened or harmed workers - although their presence was entirely foreseeable and they are covered by the Act - happen to be on the payroll of another... We do not believe that this is the result Congress intended... (under the OSHAct)."(page 9) Other issues affirmed were the violations of training under 1926.21 (b)(2) and the requiring of two bolts at connections under 1926.751 (a)."
Lessons Learned
"It is a normal procedure for the steel fabricator/erector to change the construction method of steel structures designed by the engineer, as long as the original design is maintained and engineering principles are followed. It is essential to communicate the details of the changes and the consequences of not following them (matdl.org)."
