Comcast Center

Wind-tunnel studies showed that the tower would be subject to significant wind-induced vibration due to cross-wind effects. Analysis by the structural engineers on the project, Thornton Tomasetti, indicated the building met design requirements for strength, but that wind-induced motion would exceed design criteria for comfort during high winds. Also, unless the building was significantly stiffened, under high winds structural drift would exceed design limits.

One way to reduce the wind-induced vibrations would be to increase the building’s stiffness. But that would require more structural material--and expense--than required to simply ensure the building had adequate strength for the design wind loads. Thornton Tomasetti identified a more efficient and robust solution: incorporate a supplementary damping system to reduce movement of the building without increasing overall material and cost. We developed a system that did just that. Informed by our experience developing motion- and vibration-control systems for structures around the world, we designed a unique supplementary damping system that, while demanding minimal space itself, would ensure occupant comfort on the upper floors – the most valuable rentable space of the development.

We executed our analysis and design in three discrete phases.

Selecting the right damping system

Preliminary dynamic analysis found that a tuned liquid column damper (TLCD) would be the most efficient way to manage wind-induced movement: it would be cost-effective and demand minimal space. Essentially a set of carefully designed water tanks situated at the top of the structure, the TLCD is a passive device: as a tower sways in one direction, the water in the TLCD naturally sloshes out of phase; the water within the tank is “tuned” to oscillate at a frequency very close to the natural frequency of the building.

This design allows the resonant transfer of energy from the building to the 1,300 tons (over 300,000 gallons) of water in the TLCD. However, without a means of removing this energy from the TLCD, it would soon resonate back into the building. To prevent this action, a series of louvers or paddles are therefore installed inside the water tanks. As water flows around these obstructions it creates turbulence, which eventually becomes heat. Due to the large thermal capacity of water, this heat is harmlessly absorbed and later passed through to the environment.

Using our unique tools to optimize the system

Most published analytical formulas for TLCD design make a key simplifying assumption: that a damper’s response to a building’s movement increases in a linear fashion. That is, if a tall tower moves twice as much, the damper will also move twice as much. Reality does not match this assumption. Dampers’ responses to building movement are typically significantly non-linear.

We’ve created a set of software tools for modelling such complex, non-linear behaviour. Instead of relying on assumptions about TLCD behaviour, we have observed them and used our measurements to develop a proprietary suite of analytical and modelling software, which can accurately predict their behavior under various conditions. Using this software, we determined the optimal final form of the Comcast Center system, testing numerous sizes and geometric configurations and eventually determining the best and most efficient in terms of both cost and space.

Refining the system’s performance on-site (with help from evolution)

After construction and installation were complete, we set to work ensuring that the damping system functioned as designed. This involved three key activities. First, we tuned the system, measuring the ambient vibration in the building to determine its natural frequencies and adjusting the TLCD’s water levels to ensure optimal response. Second, we performed commissioning tests: we tested the louver blades in various positions, measuring water levels and building acceleration in each configuration, and eventually identified the best arrangement.

A third aspect of our work to optimize the TLCD’s performance relied on a unique RWDI software tool that mimics biological evolution. Some aspects of the dynamic interaction between a building and the TLCD movement cannot be measured directly. As a result, it is difficult to get an accurate picture of how the system is performing in the real world as compared to how it was designed to perform. Measurement data gives us a partial picture of how the damper is behaving, but in order to gain a detailed and comprehensive understanding the damper’s performance--and thus be able to optimize it fully--we need a tool to fill in gaps in our knowledge where direct measurement is impossible.

We developed this tool, in the form of an algorithm that generates hundreds of possible “fits” between the motion we can actually observe and the predicted interaction of a building and a damper. Each fit represents an individual possible solution: a single explanation of the relationship between the damper’s performance and the observed data. The algorithm then selects from the population of fits “parents” that most closely agree with the observed motion. It combines pairs of parents to produce a next generation of slightly better fits. Eventually, the software evolves a “child” that represents a best fit: a solution that conforms most closely to the observed data, and fills in gaps in our understanding of the interaction between the building and the TLCD.

The TLCD is acknowledged as the largest damper of its kind in the world. Its unusual shape minimized its impact on usable floor space without compromising performance. In addition to the substantial up-front cost savings it delivered by enabling a supertall design that limited building sway without requiring additional structural material, the system will continue to enhance the Comcast Center’s performance--in terms of human comfort, diminished structural drift and rentable square footage--throughout the building’s life.