A couple of years ago I wrote a series of blog posts on incompatibility, how trying to achieve one goal in design makes another goal difficult or impossible to reach. (I suggest using the search box on the right, as those posts were not on consecutive days.) The picture above, from a factory being converted to offices, shows a milder case of this problem: all of the design goals have been achieved but with some fudging. The most important statement in this blog post: I’m not criticizing fudging and I’m not criticizing the result. That said, I think it’s important to understand what the goals were and what’s been fudged.
The building was constructed in 1928, at a time when reinforced-concrete flat-slab frames were well established. They were a nation-wide building type, and I’ve seen nearly-identical examples in cities all over. The slabs were designed as spanning two ways, using a method similar to that still in the ACI code today. The column capitals (the flared circular-cross-section areas) and drop panels (the square areas) nicely visible here served a specific purpose. The capitals pushed the shear plane (actually a truncated cone, but whatever) further out from the column centerline and thereby increased its area. This was a simple way to reduce the stress for punch-through shear, which at its worst leads to the nightmare of the slabs pancaking. The square drop panels are a little more complicated. They increase the area of concrete resisting punch-through shear, but they also, somewhat, reduce the bending stress for the negative moments at the column. All of this was needed because the live loads in these factories were high and the concrete was, by modern standards, quite weak.
If you look closely, you can see that the edges of the top panel are chamfered. This was done to help get the forms off the concrete without damaging it and to reduce damage in fires. The little triangular prisms of concrete that do not exist because of the chamfers would be fragile and structurally more or less useless. But it’s important to note that the contribution of the drop panel in design is based on the full dimension of the panel, which includes the chamfered area. That area is slightly weaker than the design assumption, but not enough to matter.
The little rectangular holes regularly spaced in the slab underside are inserts meant for hanging lightweight objects. (Probably the original light fixtures.) Again, similar details still exist. Those inserts technically interfere with force transfer within the slab but, again, the effect is small and ignored in design.
How many little things like that are there in a design? Probably dozens. Big deviations are included in designs. One of the first times I faced a design problem that didn’t look the ones I saw in school was at 712 Fifth Avenue, where I had to design a panel of slab where there were a number of pipes. The holes meant that slab did not meet the requirements for a two-way slab panel, so I ended up designing the pieces of concrete running between the pipes as little beams. Put enough little beams together and you have a slab, sort of. So that area was designed for the deviations. But areas where there was only one small pipe were not. Safety factor serve multiple purposes, and one of them is to account for all the little deviations that are too small to require design but which have a non-zero effect on strength.
