That odd photo is a brick pier in the cellar of an uptown brownstone. There’s some water damage at the base from rising damp, but otherwise no reason it couldn’t perform its structural role. The line of piers and girders in the cellar is below the stairhall partition on the floors above, which is an interior wood-framed bearing wall that cuts the floor joist span from 20 feet down to about 14 feet.

Two minor points that may make this picture easier to read. First, there’s a tin ceiling that hides the joists, with only the wood girder above the pier extending below the metal. Second, yes, the pier is both tilted and slightly bent.

You’ll have to look closely to find the interesting thing about this pier – the reason I took the picture – as it’s both subtle and partially hidden. If you just below the pipe running left to right across the top of the shot, you’ll see a gap between the brick and the wood girder it supports. It’s not a tiny gap either – it’s close to half an inch. Horizontal gaps and cracks are a little unnerving as they seem to imply that gravity isn’t working properly.

There are several reasons that the gap might exist. The partition above might be acting as a very deep beam and spanning past this pier. This pier might have settled relative to the rest of the structure, and the building is currently unoccupied and mostly cleaned out, so the live load above is quite small. Someone might have knocked into the pier hard, causing the bend and tilt as well as the gap. I only have evidence of the ten minutes I was in the cellar, a single still of a 130-year-long movie.

The most interesting thing about that gap is that it’s almost certainly not constant over time. If the floors above were fully loaded, I would expect that the structure above would settle down onto the pier and thereby gain another support. This kind of behavior, where the structure changes depending on the state of load, is familiar to engineers from trusses and cross-bracing, where slender diagonals work well in tension but are may be completely excluded from the analysis models when they are in compression. It’s less familiar to see this kind of behavior in a traditional wood-and-masonry structure, but here’s a picture of it.

Also, one of the reasons we added the Research page is to keep topics that interest us alive. The paper “Analysis of an 1864 Long-Span Truss Roof” linked there addresses this kind of structural action that changes with load.

Scroll to Top