FLOOR SPECIMEN. 17-ft-long assembly with 1/2-in. fireproofing, before the Aug. 25 test.

Recent laboratory fire tests performed on trussed floor assemblies built to mimic those used in the twin 110-story towers of the World Trade Center provided two surprises, say sources. One is that an "unrestrained" assembly performed better than a restrained one. The other is that performance of the floor system in the test "appears to be better than anticipated" because in all cases, the assemblies continued to carry load after they reached high temperatures.

The tests, conducted during August on 35-ft-long and 17-ft-long specimens with either ¹/₂ -in. or ³/₄ -in.-thick sprayed-on fireproofing, showed the assemblies were able to withstand standard fire conditions for between one and two hours. The tests were performed by Underwriters Laboratories Inc. under a $236,635 contract and sponsored by the National Institute of Standards & Technology as part of the $16-million WTC building and fire safety investigation.

At the final test, held Aug. 25 at Northbrook, Ill., NIST’s lead investigator stressed to attendees that the tests provide only a means for evaluating the relative fire-resistance rating of the floor systems under standard fire conditions and according to accepted test procedures. "The goal is to improve the test method," said S. Shyam Sunder.

advertisement ...

The tests did not attempt to mimic fire conditions on Sept. 11. They did not account for the fireproofing thicknesses at the impact floors of the north tower, which had been upgraded to a specified 1.5 in. and installed to 2.5 in. They did not account for the structurally compromised frame and floors or any fireproofing that might have been knocked off the remaining steel. They also did not test the connections, which at least one attendee thought would be valuable. "These tests alone cannot be used to determine the actual performance of the floor systems in the collapse" of the twin towers, Sunder cautioned.

Though the assemblies deflected, in all cases they continued to carry load. That might help explain the inward bowing of perimeter columns before the collapse, said Sunder. He added, however, that some floors appeared to be hanging as a result of structural damage from the planes’ impacts.

The biggest surprise was the result of the tests done on 35-ft-long assemblies. The restrained assembly deflected 14 in., mostly in the center of the span. The unrestrained assembly deflected 13.5 in., over the span.

Code consultant David Collins of The Preview Group, Cincinnati, said the superior performance of the unrestrained assembly "flies in the face" of an assumption that unrestrained assemblies need more fireproofing. Others agreed that the results will force a reconsideration of the fireproofing requirements for the two types of assemblies in building codes and standards.