Reply #22: No, that's not what I said. The curtain walls were ATTACHED to the columns [View All]

[William Seger]

When the interior beams broke away from the exterior columns, those columns could not fall in different directions as they did in the FEA model, because the curtain walls were still attached. The curtain walls apparently were rigid enough to hold themselves and the exterior columns in the general box shape as they all fell together. For example, in the model, you can see the columns fall inward at their tops -- the shorter end "walls" in particular -- deforming the "wall" of columns at the corners first, and then the deformation propagates down the walls. In the real building, it appears that the curtain walls were stiff enough to prevent that -- each wall bracing the adjacent wall from falling inward -- so the box shape was largely maintained (at least in the upper half of the building that the videos show). In the model, the rigidity of the beam and column connections was the only thing preventing the box shape from deforming.

No, NIST doesn't say anything at all about the curtain walls, neither in any of the detailed descriptions of the model elements nor in any description of the collapse. And as I already pointed out, they're not shown in any of the model renderings. Presumably that was because the curtain walls played no part in causing or avoiding the collapse, and the model had already required some simplification to get down to being 3 million elements. The curtain walls (and possibly some other simplifications) would, however, affect how the collapse looked when compared to the videos, but the model was intended only to study the overall structural response of the steel framing and the concrete floors. There is no reason to expect the FEA model to look exactly like the collapse videos if it didn't include those curtain walls.