Carillon Redux: Bell Chamber

Transmission wires and bells leave little room to maneuver inside the bell chamber.

The Hopeman Carillon has 50 bells. The largest, nearly 41 inches in diameter, weighs 1,411 pounds. The smallest, about 7 inches wide, weighs a mere 26 pounds. The three largest bells play G3, A3#, and C4, and the remaining 47 are arranged on a continuous chromatic scale from D4 through C8 – four octaves. The 50 bells are mounted on a triangular framework with the smallest bells at the top, an unusual arrangement.

The bells are fixed in place and don’t move. Instead, they are struck by clappers pulled by wires strung across the framework and down through the floor. A system of bell cranks, harnesses, and an umbrella rack keeps the wires in place and prevents water from leaking through the holes in the floor.

If this all sounds a bit involved, well, that’s because it is. Carillons share with pipe organs the distinction of being the largest musical instruments in the world, and they are just as complicated.

The first known carillon dates from 1510 in Flanders, and for centuries northern France, Belgium, and Holland remained the center of carillon development and performance. If you want to learn more, here is a brief (and well-researched) history.

My plan is to include as much of the mechanical system as possible, since we don’t yet know what will need to be shown in the final informational graphic. Details such as the clappers and bell cranks – that might normally be roughed in as low-poly shapes – will be built as high-poly meshes so they can be rendered at short distances. This will require additional modeling time now and longer render times later. Since our goal is to create a few high-resolution still images, this should be time well spent.

Floodlight model, ready for installation.
First, some tidying up

While going through some reference photos I discovered that each of the stone owls mounted on the upper dome has a floodlight attached to its backside. The lights aren’t visible from below and the angle is perfect for providing dramatic illumination of the lantern at night. The shape is simple and can be modeled quickly and attached to our owls. Why not?

Also, the university recently (and generously) provided some finely detailed, small-scale architect’s plans of the library tower, dome and lantern. Some of these include actual measurements. I’d been working from a large-scale elevation plan and had the nagging feeling that the proportions of the lantern model weren’t quite right – it looked a bit squat.

Orthogonal view of the lantern model before adjustment (left) and after.

Turns out my suspicions were right. Once the new plans were placed and (very carefully) scaled in Maya, it was apparent that the lantern model was several inches short and a few inches too wide. With a little judicious scaling and other tweaks, the problem is fixed.

Which is a good thing, because the inner dimensions of the lantern must be spot on for everything to fit. As the photo at the top of this post shows, it’s crowded in there.

Horizontal sections indicate positions of the bells and console. (Courtesy of University of Rochester Libraries/Department of Rare Books, Special Collections, and Preservation)
Building the frame

In addition to the bell chamber elevation diagram, we also have three horizontal sections that show the top and bottom rows of bells, and the position of the console.

Horizontal sections are scaled and moved into position.

These are placed in Maya and aligned with the lantern model. Happily, everything matches perfectly.

Four upright I-beams support the carillon.

Hiding the lantern and dome gives us a clear space to work. Four upright I-beams, which support the carillon’s 6,668-pound weight, are set in place.

Internal structures are added to the frame.

Horizontal supports, service platforms, and the maintenance ladder are added. Space is already filling up . . .

Bell chamber framework with bell supports.

And the eight triangular frames that will support the bells are modeled and moved into place, along with assorted brackets and bolts, using the elevation plan as a guide.

Maya’s curve tool is used to trace the largest bell’s profile.
If I were a bell

We know the diameters and note frequencies of each of the 50 bells, but not the heights or precise profiles. The best bell diagram I have is the contractor’s elevation plan, which includes sketches of each row. Using this, we can trace the profile of the largest bell – bell 1 – with Maya’s curve tool, while guessing the shape of the interior surface.

Traced curve in Maya (left), and revolved bell NURBS surface.

Then, as we did with the lantern cupola and candlesticks, we can revolve the curve to create the bell shape.

Bell no. 1 is moved into place.

The bell is moved into position in the framework, just to see if it fits. (It does.) Only 49 to go.