Forty two concrete tubes packed side by side on the Cape Town waterfront, eighteen feet across and a hundred and eight feet tall.
They held maize for most of a century. Then they held nothing.
The walls are seven inches thick, which is thin for something that stands 187 feet.
A tube that shape survives because it is round. Cut a hole in it and the geometry that was holding it up is gone.
Somebody wanted a room in the middle of that.
The answer was not to take concrete out. It was to put more in.
Why thin round tubes resist being opened
A silo cell works in hoop tension. Grain pushes outward, the ring of concrete pulls against itself all the way around, and nothing bends.
The wall can stay thin because it never has to act as a beam. It only has to hold a circle.
Cut an opening and the circle stops being a circle. The load goes looking for another path and there is not much steel in there to find one.
That was the situation here. Old concrete, sparse reinforcement, and a proposal to remove the middle of the block.
So the sequence got inverted. Crews cast a new reinforced lining inside the tubes first, bonded to the old wall.
Seven inches became about sixteen inches of composite, and only after that did anybody start cutting anything.
What the inside of the block was like
The tubes are not rooms. They are shafts, closed at the bottom, with no floors and no windows.
Standing in one before the work you would see a circle of daylight a hundred feet overhead and dust on every surface.
The complex was finished in the mid 1920s and graded maize brought in from across the country.
It came out of service at the start of this century and then sat, too solid to fall down and too awkward to use.
For years the standing question was what anybody could possibly do with a block of sealed concrete pipes.
Every conventional answer ran into the same wall, which was that the interior had no usable volume.
The cut itself and the numbers behind it
The void in the middle was not drawn freehand. It was taken from the shape of a grain of corn and scaled up until it filled about 89 feet of height.
That surface was broken into thousands of coordinates and marked on the concrete physically, nail by nail, so the saw crews had a line to follow.
Diamond saws did the cutting, working down through walls that had been thickened for exactly this.
Roughly 160,000 cubic feet of concrete came out of the middle of the block.
The cut faces were then polished, so the exposed edges read as a smooth surface rather than a demolition scar.
The finished void is a negative shape, and everything around it is the original structure.
Why nobody just knocked it down
Demolition was the cheaper line on paper and it was never taken.
Part of that is heritage. The grain silo was the tallest thing in the city when it went up, built between 1921 and 1924 by indentured labor, and it is the landmark the waterfront grew around.
Part of it is arithmetic. Clearing 42 tubes of dense concrete on a working harbor front is not cheap either.
Reuse also kept the outline of the waterfront intact, which nothing built new on that footprint would have done.
The conversion came in around 37 million dollars and produced about 64,600 square feet of exhibition space across nine floors and roughly 80 rooms, opening in September of 2017.
The engineering write up of the composite sleeve approach is in the project record, and it is the part that made the rest possible.
The construction sequence, including the sawing, is set out in a build account that puts the added lining at about ten inches on its own.
What the method is actually good for
Adding structure before subtracting it is not specific to silos.
It applies wherever an old element is strong in one direction only and somebody wants an opening through it.
The failure mode when that step is skipped is not subtle, as it was not when a backpropping failure ran 105 feet across a slab under a fresh pour.
Concrete is also getting better at looking after itself, with embedded bacteria sealing cracks that would otherwise need a crew.
None of that helps a structure already standing, which is why the sleeve trick still matters.
The lesson is dull and useful. Old concrete is not weak so much as single minded, and the work is teaching it a second job.
