Observed Damage - Main Span - Cross Section

At the Pylon/Main Span connection, there were many problems. Concrete crushing and spalling was observed on the under side of the deck. The damage seemed to be more pronounced on the side of the missing wings. This follows logic given the stiffness discontinuity when bending in the direction where the missing wings would have been in compression. The damage seen would be completely consistent with simple week/strong axis bending failure were it not for the large area of crushing and spalling along the length of the bridge spine. This longitudinal crack is a source of on going investigation.

	Figure 1 - Under the deck on the south side of the Pylon (P13 side). Damage is nearly consistent with bending failure.

Inside the box girder, longitudinal bars buckled as the lower slab was compressed in and then dropped (Figure 2). Also worth noting, was a horizontal splitting crack located near the middle of the center web of the girder.

	Figure 2 - Longitudinal steel buckled as the bottom slab of the box girder separated under compression failure

Moving closer to the outside, the lower slab dropped even more, meaning there was a component of twist about the central axis of the spine. With the increased vertical deflection in the lower slab, and presumable larger high strain cycles, some of the longitudinal bars were fractured.

	Figure 3 - Bars buckled and fractured more towards the outside of the spine and the vertical deflections were more pronounced

Moving up the outside web on the Eastern face of the box girder, heavy compression damage observed. Longitudinal bars buckled under a seemingly uniform compression field (Figure 4). This idea would support a failure mode of bending in the strong axis.

	Figure 4 - a nearly vertical line is seen on the spalling boundary on the outer web coupled with longitudinal bar buckling

All along the inside of the outside web, the transverse reinforcement was bowed out and the concrete was pulverized into rubble (Figure 5). The vertical line at the spalling boundary was seen to extend the entire height of the outer web. This line could represent a line of constant strain for support in an argument of bending failure.

	Figure 5 - Transverse bars bowed heavily as they were pushed by buckling longitudinal reinforcing

Moving now to the outside of the Eastern face (the area were the wings had not yet been placed), we see the other side of the damaged outer web (Figure 6). There is evidence of inward motion of the end of the web where the web crushing occurred. Furthermore, there is distinct shear failure seen here in the lower slab as the slab was dislocated and shoved forward into the solid concrete diaphragm at the pylon. If the failure mode was indeed bending about the strong axis, as signified by the line of constant strain, then the bottom web shear failure seen here could be called a local buckle.

	Figure 6 - Outside of the Eastern face of the main span showing compression failure in the web and shear failure in the lower flange

The under side of the top slab showed a higher mode of bar buckling along with a permanent downward slant (Figure 7). This is consistent with a local axial load failure that could be explained by a top in compression weak axis bending overload..

	Figure 7 - Inside the box girder looking up

From the top of the deck, the slab appeared to have failed in compression in the two perpendicular directions (Figure 8). The deck failure transverse to the longitudinal axis of the bridge could be explained by the overload caused from the inverted V prestressing. As the dead load is taken off the structure from the earthquake, the top of the slab may have been overloaded in compression. This is an area of research focus.

	Figure 8 - Compression failure seen in both directions on the deck surface

From the top of the deck, a five inch vertical offset (Figure 9) was seen on the East side. There was no vertical offset measured on the West. This fact suggests a twisting motion resulted from the section failure.

	Figure 9 - A five inch vertical offset was measured on the East side of the box girder deck surface.

On the West side of the center web, a shear crack observed. The direction of the shear crack suggests an upward vertical force overload along the bridge span. Out of plane bending of the tower causing the cables to generate upward force resulting in this overload is hypothesis currently being investigated.

	Figure 10 - Web shear crack suggesting a upward vertical loading along the span