Evolving approaches to Maintenance and the Arecibo Radio Telescope

https://www.ucf.edu/news/broken-cable-damages-arecibo-observatory/

"One of the auxiliary cables that helps support a metal platform in place above the Arecibo Observatory in Puerto Rico, broke on Monday (Aug. 10) causing a 100-foot-long gash on the telescope’s reflector dish. Operations at the UCF-managed observatory are stopped until repairs can be made."

While this is a terrible thing to happen to such an important facility and scientific research tool, it has generated some interesting discussion with respect to design and maintenance practices and how approaches to managing the life of equipment capabilities have changed in the last 60 years.

My thoughts below aren’t intended to identify the cause of the cable failure as I am not a member of their maintenance or support team or privy to their maintenance concept, but rather use the event as a means to generate discussion.

Being the ILS/LSA geek that I am, one of the first thoughts that went through my head as I read the article was “I’d be really interested to see the results from the SSI (Structurally Significant Item) logic for RCM (Reliability Centered Maintenance) the design team did”. That’s when the next thought hit me “…hang on a minute, when was this thing built again?”

The original Arecibo concept was derived from early 1950's anti-ballistic missile tracking studies and adapted as requirements changed, as they tended to do rapidly in the post-war, pre-cold war world. Eventually the design was modified and engineered to take advantage of natural geological formations found on the island of Puerto Rico, eminently suited to housing large spherical reflectors. The facility was constructed between 1960 and 1963 which implies that there are likely key structural components that are around 50-60 years old. Many people will argue that there are both civilian and military airframes still flying safely today that are of a similar age and I will agree, but they attract a high degree of safety critical maintenance inspections and practices as part of certification and airworthiness requirements.

If we consider the formal approach for development of a maintenance program that many of us understand nowadays, loosely shown as: Concept - Design - FMECA - RCM - MTA - Maintenance Program, much of this was simply not available or even "invented" as a formalised, documented methodology at the time the radio telescope was initially designed and built.

Failure mode, effects, and criticality analysis (FMECA) had been developed and formalised as MIL–P–1629 by 1949, so you can assume that elements of it were used to iron out significant safety and catastrophic (high likelihood) failure modes from the design. Reliability Centered Maintenance (RCM) however, would not have been a consideration, as Nowland and Heap were only beginning to develop their methodology in the 1960's and wouldn’t even publish it as RCM until 1978. RCM itself was built around the previous work that had been done by the Air Transport Association, published as the Maintenance Steering Group - MSG-1 (1968) and MSG-2 (1970), so an aviation industry targeted approach.

If the maintenance practices of the time follow their usual approach, it would likely have comprised a mix of scheduled overhauls/rebuilds of key components and sheer over-engineering to mitigate likelihood of fatigue related failures occurring. This goes counter to our approach these days of “designing out” as many high likelihood / severe criticality failure modes as possible and mitigating anything else that remains through scheduled maintenance inspections with some life-limited (structurally significant) replacement parts (as required). Obviously, the modern approach results in a higher degree of availability, less corrective maintenance and lower maintenance costs through in-service inspection routines and not costly “overhauling for the sake of overhauling”; being one of the key drivers behind the development of RCM.

This is not to say that the facility hasn’t gone through upgrades and modification programs through its life and likely attracted a rigorous FMECA/RCM approach for those new systems. It does however remain that the entire framework and design it was built on did not have all of these analyses available and there will continue to be some structural elements that have inherent failure modes that cannot be fully mitigated through RCM practices or even an Age Exploration program; they will eventually hit a limit and fail.

Design for maintainability and reliability centered maintenance practices – concepts we consider as par for the course nowadays but tend to forget that they are relatively new when held up to older capabilities like the Arecibo Radio Telescope, designed in the 50’s and built in the 60’s.

It would be an interesting case study:

• What sort of drone or remote-assisted NDI / inspection procedure could be conducted on a 60yo facility that's 100% exposed to the elements?
• Can a drone conduct remote inspections without causing EMI issues during facility uptime?
• Are there / were there hard-life, fatigue related lifing parameters for 3-inch auxiliary support cables, and is the definition of “auxiliary” in this case "backup or redundant?" (and the leading questions that raises).
• Then consider the facility’s remoteness (1000km from mainland USA).
• The availability of parts.
• Maintenance training.
• Available onsite maintenance crew.
• Operating concept defining the expected uptime and system availability and the follow-on identification of the time remaining to conduct maintenance at all.
• Are there compounding structural maintenance issues arising from the post-Hurricane Maria repairs that are still underway?

Like I said, it’s an interesting discussion and one I’m sure to expand on and use when we deliver future ILS / LSA courses and discuss the impact of changing maintenance practices.