Suspended Structure Collapse: Arecibo Observatory

Arecibo

BY: Bob Glenn, Editor and Publisher

[This article originally appeared in the September/October 2024 Issue of Wire Rope Exchange. See the entire issue online at this link: September/October 2024 Wire Rope Exchange.}

[Photo Above: Using work vehicles and equipment, the cleanup team removed debris and components of the platform that fell onto the dish and down the mountainside. Photo Credit: Jacobs]

Upon its completion in 1963, Puerto Rico’s Arecibo Observatory became the world’s largest radio telescope – and remained so for most of its working life. The reflector dish, 1,000 feet in diameter, was built into a natural bowl. Overhead, three concrete towers and more than 21,000 feet of steel cable suspended an antenna on a 1.8 million pound platform

Under the guidance of the National Science Foundation (NSF), an operating organization led by the University of Central Florida (UCF) manages the facility. It is credited with several “firsts” including discoveries of a planet outside of our solar system and a binary pulsar. It’s also been important to the study of the ionosphere and its composition, and tracking the orbits of potentially hazardous asteroids, among many other accomplishments.

A series of three cable failures in 2020 brought the radio telescope’s use to an end. Ironically, the 1995 James Bond film Goldeneye foreshadowed the telescope’s fate when a Bond villain ostensibly falls into the dish and the assembly – in flames, of course – crashes down onto him.

Platform structure of the Arecibo Observatory taken August 3, 2020, one week before the initial cable failure. Photo Credit: University of Central Florida

The first failure occurred on August 10, 2020, and was followed by another on November 6 of that year. On November 19, the NSF accepted a recommendation from Thornton Tomasetti, a global specialist engineering firm, to decommission rather than repair the facility. Shortly after, the mission shifted from decommissioning to salvage when a third cable failure caused the suspended structure to collapse into the dish below.

All of the failures occurred during fair weather at the points where zinc-filled spelter sockets connected cable ends to the support structures.

After the initial cable failure, UCF launched an analysis of the socket connection and cable end at the heart of the that failure with the intention of determining appropriate repair and remediation measures. A report, prepared by Wiss, Janney, Elstner Associates (WJE), was incorporated into subsequent investigation.

After the total collapse, Thornton Tomasetti was retained to perform a forensic analysis on all of the cable failures. Their work involved a number of other significant contributors in addition to the WJE report. They include Socatec (a specialist in Testing, Inspection and Certification), Lehigh University’s Fritz Laboratory, Dr. Adrian Brugger of Columbia University, the Oak Ridge National Laboratory, and Langan (environmental engineering consultants).

The cleanup team was on site shortly after the collapse to assess the damage. The team developed safe and environmentally sound plans to complete the cleanup. Photo Credit: Jacobs

The analysis included a full-scale load testing and neutron radiography of one or more of the sockets involved, as well as an exploration of the earthquake resilience of the structure, as it had been subject to that stress in the past.

The Incident Report

While Thornton Tomasetti focused most specifically on the sockets where the failures had occurred, they did so in context of the complete structure. They reviewed the initial design and examined two major upgrades made in subsequent years. They noted that the structure had been designed with hurricane impacts in mind, but that the overall structure essentially had a 2-to-1 safety factor with respect to force carried by the cables, which they considered relatively low for a cable-supported structure.

They also explored maintenance history, which included remediation of paint deterioration, intrusion of moisture and individual wire breaks. Analysis concludes that these issues had been appropriately addressed.

In the reflector area, debris removal operations were coming to a close in September 2021. Photo Credit: Thornton Tomasetti

Thorton Tomasetti’s report noted that a 2003 inspection found that cables had slipped by as much as half an inch out of their sockets – and a subsequent inspection 8 years later documented that the degree of cable slip remained unchanged. The initial inspection report stated that the cable slip observed dated back to fabrication and testing. Thorton Tomasetti was unable to document that facility staff had been directed to periodically inspect cable slip, nor was there anything to indicate that a maximum value for allowable cable slip had been communicated to Observatory staff.

In 2017, Hurricane Maria battered Puerto Rico, making landfall with 155 mile per hour winds and spreading hurricane force winds over much of the island. The incident report indicates that Observatory inspectors found cable slips of more than one inch, but there was nothing showing that these was directly associated with the hurricane, and nothing to indicate that this was thought to threaten the integrity of the structure.

In laboratory analysis of the cables and sockets, the cable was found to be in good condition, and met parameters for material composition and strength called for in the original design.

The sockets were another matter. Thornton Tomasetti reviewed the WJE report on the first socket failure, and dissected five more – including one of the other failed sockets and another that showed the most cable slip. This put focus on the zinc castings, which inside the more distressed or failed sockets show “significant material flow” which enabled slippage of the cable. They also noted that the mechanical properties of the zinc varied within each socket as well as between sockets.

This view shows the damage to the rim wall, the damaged platform, and fallen panels from the reflector dish underneath the platform. Photo Credit: Thornton Tomasetti

Analysis of the overall structure of the telescope under a range of conditions it might be expected to encounter – operational and environmental – led the investigation to state that “the cable system was highly stressed under its own weight” but in a manner consistent with the original design parameters. That said, extreme events – the hurricane in 2017 and an series of earthquakes in early 2020 – may have increased tension by as much as 15 percent and could have extended the degree of cable slip in the sockets.

The sequential nature of the failures was also subject to this investigation. The initial failure was an isolated cable installed more than 20 years before the failure. Its failure would have increased stresses through the remaining cables, and also causes the suspended platform to rotate.

Finally, investigators wanted to understand why – under similar loads and stress – some sockets exhibited a large degree of cable slip, and failed, while others did not. In addition to other analysis, a load test to failure was performed on one of the sockets. The investigators determined that the critical factor was splaying of strands in the zinc during placement of the socket. To quote directly from their report:

“Zinc slowly flows under shear stress when the splayed-out wires, which are typically the two outer layers of the cable, cannot alone resist the cable tension. This condition exists when the wires are insufficiently splayed out for the given cable tension. In this condition, the socket is bound to fail because the zinc will continue to flow until the cable pulls out of the socket entirely.”

They further conclude that this condition is not likely to be revealed by proof testing conducted for minutes or even hours. Instead, this condition plays out over an extended time, revealing its existence through observable cable slip.

A failed cable and socket recovered from the debris. Some of the cables and sockets were tested and evaluated to help determine the cause of the collapse. Photo Credit: Thornton Tomasetti

Based on that analysis, investigators concluded that five factors led to the demise of the facility, stated as follows in the executive summary of their report:

(1) the manual and inconsistent splay of the wires during cable socketing
(2) the design of the cable system with relatively low safety factors
(3) the occurrence of extreme environmental events such as hurricanes and earthquakes
(4) the non-replacement, repair or bypass of the sockets where large cable slips were observed
(5) the addition of auxiliary cables as isolated cables

Looking forward, the report cited essentially two primary factors for consideration in similar future installations. The first was better quality control in the socketing process with respect to the number and geometry of splayed strands. The second was to suggest that a larger than 2-to-1 safety factor should be supplied for similar structures.

The Future of Arecibo

After considerable work to clear debris from the site and rehabilitate ancillary structures, the site now hosts the NSF Arecibo C3 STEM Center, which in 2024 began a series of pilot programs for promotion of science, technology, engineering and math rooted in the legacy of the site. The “C3” is a reference to Ciencia (Science), Computación (Computing), and  Comunidad (Community).

Visit www.thorntontomasetti.com/search/Arecibo to access their complete forensic report and more details about their analysis of the Arecibo Observatory incident. Visit nsf.gov to learn about the National Science Foundation and search “Arecibo” for more about that facility.

Get the free newsletter

Sign up for news, trends and equipment for sale in the lifting and rigging industry.
By signing up you agree to our Terms of Use and Privacy Policy. You may unsubscribe at any time.

Latest Products For Sale

Type Description
Lifting Products
Hooks
1/2” Grade 70 clevis grab hooks
Lifting Products
Chains
1/2” Grade 70 chain 13 drums 200’ each (Make Offer)
Lifting Products
Hooks
3/8” Grade 70 clevis grab hooks
Lifting Products
Chains
3/8” Grade 70 chain 6 drums 400’ each (Make Offer)
Equipment Used 75’ pull test bed. Currently in 3 pcs. 25’ long would need to be…

News & Articles