5

Other Lessons Learned

STATE OF KNOWLEDGE

Transfer of Institutional Knowledge

The Arecibo Telescope is owned by the National Science Foundation (NSF) and was managed by Cornell University from its opening in 1963 to 2011. Management was transferred to a consortium led by SRI International in 2011 and to another consortium led by the University of Central Florida (UCF) in 2018.¹

Cornell University had to make changes in leadership, staff, and administration working on the Arecibo Telescope for nearly 50 years. How much knowledge and experience with the design and performance of the telescope structure was transferred both during internal organizational changes and later to SRI International and subsequently UCF is unclear. Certain performance aspects, such as temperature, oscillations, wire breaks, and tiedown stresses of the telescope, were monitored. Ultimately, these metrics alone do not paint a complete picture of the cause of the failure.

Hurricane Irma occurred on September 7, 2017, and Hurricane Maria was on September 20, 2017. The transition to the new management under UCF in the first half of 2018 happened at a critical time in the structure’s history, both in terms of funding and the impact of the hurricanes on the structure.

It seems unlikely that UCF had adequate time and resources to review and understand the Arecibo Telescope’s original 1963 design, the 1974 upgrade, the 1997 upgrade, the structural inspection and maintenance records produced for nearly 50 years, the performance over time, the critical aspects, and the key factors, such as the wire breaks and cable pullout of the sockets and their significance on the strength and integrity of the structure.

In a 2003 inspection, the telescope’s Engineer of Record observed that cables had slipped out of their sockets by as much as one-half of an inch and attributed the slips to have occurred during the fabrication or testing of the cables. A later report in 2011 noted that the cable slips remained unchanged from 2003. To our knowledge, the Observatory staff was not instructed to monitor cable slips, nor were they provided a limit on acceptable cable slip. After Hurricane Maria in 2017, the Observatory staff observed and recorded cable slips of more than one inch on two of the sockets.

There is no documentation to show whether these cable slips increased during the hurricane, and to our knowledge, they were not identified as an immediate structural integrity issue.²

From this record, one may conclude that the telescope’s Engineer of Record, the Observatory staff, and the staff of UCF did not recognize the significance of the cable pullout. The measured cable pullout may have appeared “normal” to them and was not on their radar as signs of structural distress. The lack of concern may be because a small cable pullout was present from the beginning, and no one in authority had previously raised an alarm:

• February 19, 2019: A pullout of 1.125 inches of the cable was measured at the location of M4N-T. However, it would seem that nothing was done about this excessive pullout.
• August 10, 2020: First socket failure. About 18 months after the excessive pullout of the cable was measured, the cable pulled entirely out of the socket at M4N-T.

Finding: Record keeping related to telescope organizational and structural performance was often inconsistent and incomplete over time, including through operational contract transfers.

Documentation

A facility’s “owner’s manual” should be updated and passed on to other generations of owners/stakeholders and engineers-of-record of the facility. Innovative, unusual, one-of-a-kind structures, long-span, and high-rise structures, particularly, require such documentation. The documentation should provide a list of the governing design building codes and regulations, a detailed description of the structural system, the design life span, identify the critical elements of the structural system, list the strength and types of materials used, the design forces from gravity (dead and live), lateral loads from wind, earthquakes, soil, and water, the design wind speeds for strength and serviceability, the load combinations, the design programs used, modeling assumptions, the building periods, and results of various code checks. The redundancy of the system should be documented, i.e., the various alternate load paths for various scenarios and the reduced factors of safety used in conjunction with these “what if failure” mode cases.

Additionally, a list of signs of distress to be observed at regular intervals by inspectors should be provided. For example, this list might include buckling of structural members, excessive settlements of foundations, leaning of the structure, excessive stretching of tensile elements, cracking in the structure (beams, columns, slabs, trusses, etc.), cable slips, and wire breaks.

Beyond the need for regular inspections, there needs to be timely maintenance and repairs as required based on the inspection reports.

CONTINUED RESEARCH

Forensic testing was performed on only a fraction of the cables and sockets recovered at Arecibo. Additional sockets and wire sections were retained by Thornton Tomasetti, Inc., and are temporarily being stored. Additional connections with measured slippage are available as well as untested platform connections. Research could be conducted on those sockets that could be informative to design and materials science. These sections may have tremendous value to the research community in evaluating the low-current electroplasticity hypothesis proposed in the report; materials testing of multi-decade, in-service, large-diameter steel cable and zinc; or further analysis of brooming.

Recommendation: While still available, the National Science Foundation should offer the remaining socket and cable sections to the research community for continued fundamental research on large-diameter wire connections, the long-term creep behavior of zinc spelter connections, and materials science.