What a Column Free Corner Costs a Tall Building
By Michael Stern, JDS Development Group
The question arrives early and it is always put the same way. Can we get the column out of the corner. On a narrow floor plate that question is the plan rather than a detail of it, because the corner is where the rooms want glass on two sides and where the structure wants something solid.
This is written for architects, structural engineers and developers working on footprints small enough that the answer changes the building. The honest answer takes a few minutes to give properly. It starts with what the column is actually holding.
What a Corner Column Is Actually Holding
Under gravity alone the corner column is the least consequential column on the perimeter. It collects load from a quarter of a bay where an interior column collects a full one, and on that arithmetic it looks like the easiest thing on the floor to delete. Under wind it is doing something else.
A tall building resists overturning by pushing down on one side of itself and pulling up on the other, and the perimeter columns are where that couple gets resolved. A column standing where two faces meet participates in both directions at once, whichever way the wind is running that afternoon.
On a wide building that contribution is one of many. On a narrow one it is a measurable share of the whole. Whatever the corner carries has to reach the ground on a continuous path, so deleting it at the top of a building is a different proposition from deleting it at the fortieth floor, and a plan that clears the corner for some levels and not others is really two buildings stacked on each other.
The second thing a corner column does is receive the beams that frame into it. Take it away and the two perimeter beams that used to land there have to span the full corner and cantilever past it, which changes the depth of both. That is where the cost begins, and it is a cost in dimension rather than in anything visible on an elevation.
How a Column Free Corner Gets Paid For
There are three ways to pay for a corner with nothing standing in it, and a project chooses one of them whether or not anyone says so out loud. They are not interchangeable. Each one takes its payment from a different part of the building.
The first is depth at the perimeter. The spandrel beams carrying the corner grow deeper in order to span and cantilever, and that depth comes out of ceiling height or gets added to the floor to floor dimension, which raises the building for the same floor area. Raising a slender building is never free, because a taller prism collects more wind, which is a problem some towers answer by leaving a whole level open to it.
The second is a transfer structure that picks the corner load up at intervals and walks it inboard to the core or to a stronger line of columns. That needs a deep element at one level or several, usually where a mechanical floor already sits, and it buys clear corners for every level in between. The level that carries it spends itself on something no resident will ever use.
The third is to make the core do more of the lateral work so the perimeter does less. That is the quietest of the three and the most expensive in the place that matters, because core thickness is taken from usable area on every floor rather than from one. A tower as narrow as 111 West 57th Street, the tallest and most slender residential building in the Western Hemisphere, has to resolve this before a unit plan is drawn at all.
Each route is paid in a different currency. Depth costs ceiling height, transfer costs a level, and core thickness costs area everywhere. A team that knows which currency it is spending is making a decision, and a team that does not is receiving one.
A corner column that disappears from the plan reappears somewhere else in the building, as beam depth, as core thickness, or as a level given over to transfer.
Where a Competent Engineer Would Say This Is Wrong
The argument against everything above is that I am overstating what a corner contributes. In a core dominated scheme the perimeter columns are largely carrying gravity, the lateral system lives in the walls around the elevators and the stairs, and the corner is close to free to remove. Engineers who spend their time on wider plates make that argument, and on wider plates they are right.
My answer is that slenderness changes the arithmetic rather than the principle. The narrower the footprint, the smaller the lever arm a core can offer on its own, and the more the perimeter has to be recruited to make up the difference. The buildings where a corner is cheap to remove tend to be the buildings that never needed the perimeter in the first place.
There is a second objection worth taking seriously, which is that the depth I am describing is often already there for other reasons. A perimeter beam sized for a facade bracket or for a long span elsewhere on the plate may carry the corner with nothing added. Published member criteria from bodies such as the American Institute of Steel Construction are how that gets checked rather than assumed.
The compromise I accept is that a corner carried on a cantilever moves more than a corner carried on a column. The movement is small and entirely ordinary, and it still has to be absorbed at the facade in joints the panels are detailed to take, which is slow work that nobody sees from the street. Serviceability limits published by the American Concrete Institute and its equivalents are a starting point for that conversation rather than the end of it.
On more than one plan I have looked at, the right answer was to keep the column and draw the rooms around it. That is an unglamorous outcome and it is sometimes the correct one. A corner that holds a column and a plan that works is better than a clear corner and a ceiling nobody wanted.
Who Decides the Corner and When
This gets decided by the structural engineer and the architect together, before the plan is fixed and long before anyone draws a kitchen, with the facade consultant sizing the joints that absorb the movement. A developer contributes by asking the question early enough that the answer is still cheap to act on, and by being straight about which currency the project can afford to spend. Asking it late turns a design choice into a rework.
If a corner on your own plan is in question, the person to retain is a structural engineer of record who has worked at that slenderness, and the structural engineering associations are a reasonable place to begin looking. Nothing written here is engineering advice. No corner has ever been decided from a general principle.
On a narrow footprint the corner is decided before the apartment is, and every plan that follows is drawn around the answer.
Michael Stern is the founder and chief executive officer of JDS Development Group.
