The Doubled Load on Box Beam Four

How a Simple Rod Substitution Caused the Hyatt Regency Walkway Collapse

Fifteen hundred guests gathered in the four-story glass atrium of the newly opened Hyatt Regency Hotel in Kansas City, Missouri, on the evening of July 17, 1981, for a big-band dance competition. High above the lobby floor, three 120-foot concrete-and-steel pedestrian skybridges spanned the open space. The second-floor walkway hung fifteen feet above the lobby, the fourth-floor walkway hung directly above the second-floor walkway at thirty feet, and the third-floor walkway was offset to the east. At 7:05 PM, with sixty-three spectators standing on the second-floor and fourth-floor skybridges, a loud metallic pop echoed from the ceiling connections. The fourth-floor walkway tore free from its hanger rods and crashed straight down onto the second-floor walkway fifteen feet below. Together, sixty-four tons of concrete, structural steel, and plate glass slammed onto the crowded lobby floor, killing 114 people and injuring 216.

In the original structural blueprints prepared by engineering firm Gillum-Colaco in 1978, both the fourth-floor and second-floor walkways were designed to hang along six pairs of continuous, 46-foot-long, 1.25-inch-diameter threaded steel hanger rods. Those rods hung from the atrium roof trusses. Each continuous rod was meant to pass straight through a hole drilled in the fourth-floor transverse box beam, where a washer and hexagonal nut underneath supported the fourth-floor walkway. The same rod then continued fifteen feet further down to support the second-floor box beam at its lower end.

When steel fabricator Havens Steel reviewed the drawings in early 1979, shop detailers objected that 46-foot continuous threaded rods were impractical to manufacture, ship, and install. Ironworkers would have had to spin hexagonal nuts up twenty feet of exposed steel threads to reach the fourth-floor beam, risking thread damage during erection. Havens proposed a seemingly simple substitution on Shop Drawing 30: replace each single 46-foot continuous rod with two shorter offset rods. An upper rod would run from the ceiling truss down to the fourth-floor box beam. A second lower rod—offset four inches inward along the same box beam—would hang from the bottom flange of the fourth-floor box beam down to the second-floor walkway. Gillum-Colaco stamped the shop drawing approved without recalculating the connection forces.

That two-rod substitution fundamentally altered the free-body static equilibrium of the fourth-floor box beam. It was the mechanical equivalent of two climbers hanging from a cliff where the lower climber lets go of the main rope and grips the upper climber's boots instead. In the original continuous-rod design, the second-floor dead and live load $P_{\text{lower}}$ traveled straight up the steel rod to the roof truss without loading the fourth-floor beam connection:

In the modified two-rod arrangement, the lower hanger rod pulled downward with force $P_{\text{lower}}$ on the fourth-floor box beam. Consequently, the single upper hanger rod nut and washer four inches away had to carry the combined weight of both skybridges:

Each transverse box beam had been fabricated by welding two 8-inch structural steel channels (MC8x8.5) toe-to-toe along their thin outer flanges without internal stiffening plates or bearing pads. Both hanger rods passed directly through that thin central longitudinal weld seam. National Bureau of Standards laboratory testing after the collapse showed that even under the original $1P$ load of roughly $9,500\text{ pounds-force}$, the unstiffened box-beam flange achieved less than half the safety capacity required by the Kansas City Building Code. The two-rod modification doubled the downward shear load on the upper rod washer to $2P \approx 19,000\text{ pounds-force}$. That load exceeded the $18,600\text{ pound-force}$ mean ultimate punching shear strength of the weld. Under the dancers' weight, the thin steel flanges curled upward at Box Beam 9U, pulling the washer straight through the split seam and unzipping the span.