Hi and welcome to Failurology; a podcast about engineering failures. I’m your host, Nicole
And I’m Brian. And we’re both from Calgary, AB.
We have some exciting news! We’re starting a Patreon with mini failure episodes; failures that either have simple causes or not enough information for a full episode.
On to engineering news, researchers have found a way to make ultrathin surface coatings robust enough to survive scratches and dings.
The new study, completed at the University of Illinois Urbana-Champaign found that the rapid evaporative qualities of a specialized polymer containing a network of dynamic bonds in its backbone help form a water-resistant, self-healing coating of nanoscale thicknesses.
The study was focused on boosting the efficiency of steam power plants by making the condenser surfaces more water-resistant and efficient at forming water droplets, which optimizes heat transfer
In steam plants, with the heat and the humidity, thin coatings can break down in a matter of weeks, or even hours, making them impractical. And thick coatings can reduce heat transfer, even if they are more durable.
One thing that makes the thin coatings weak is that they form tiny pinholes as they cure, which is an opening for steam to penetrate and delaminate the coating
The coatings they were studying have self-healing properties and are able to respond to pinhole defects or even scratches
Applications - self-cleaning, anti-icing, anti-fogging, anti-bacterial, anti-fouling, enhanced heat exchange coatings
Now on to this week’s engineering failure; the Arecibo Telescope, a giant 1000 foot spherical reflector radio telescope located near Arecibo, Puerto Rico.
After construction began in the mid-1960s, it opened on November 1st, 1963, the world’s largest single-aperture telescope for 53 years.
Located in Puerto Rico as part of the Arecibo Observatory. Located about 100km west of San Juan.
Ran by the University of Central Florida since 2018 but had been previously operated by various science and government groups over the years
The telescope’s origins trace back to the US’s late 1950s attend to develop anti-ballistic missile defences.
Was used for research into radio astronomy, atmospheric science, radar astronomy, Search for Extra-Terrestrial Intelligence programs.
Also appeared in the James Bond film – GoldenEye and 1997 Sci-Fi movie Contact.
It has scrutinized our atmosphere from a few kilometres to a few thousand kilometres where it smoothly connects with interplanetary space. With its radar vision, it studies the properties of planets, comets and asteroids. In our Galaxy, it detects the faint pulses emitted hundreds of times per second from pulsars. And from the farthest reaches of the Universe quasars and galaxies emit radio waves which arrive at earth 100 million years later as signals so weak that they can only be detected by a giant eye like this one.
Discovered the first extrasolar planets around the pulsar B1257+12 in 1992
Produced detailed radar maps of the surface of Venus and Mercury
Discovered that Mercury rotated every 59 days instead of 99 days.
American Astronomers Russell Hulse and Joseph H. Taylor Jr. used Arecibo to discover the first binary pulsar. They showed it was losing energy through gravitational radiation at the rate predicted by Albert Einstein’s theory of general relativity. Won the Nobel Prize for physics in 1993 for their discovery.
The main collecting dish (the reflector) is a 305 m diameter spherical bowl that was constructed inside a karst sinkhole. More than three US football fields wide.
The depth of the dish is 51 m (~14 stories)
The dish surface was made of 38,778 perforated aluminum panels, each one was approximately 3 feet by 7 feet.
The individually adjustable aluminum panels are supported by a network of cables that are strung across the underlying karst sinkhole. These panels, making up what is called a Gregorian dome, were installed in 1997 to replace the original 20mm galvanized wire mesh that was originally laid on the cables.
150 metres above the reflector is a 900-ton platform, suspended on 18 cables that are strung from three reinforced concrete towers.
The reflector platform had a rotating bow-shaped track that was 93m long and called the Azimuth arm. This carried receiving antennas and secondary and tertiary reflectors.
Each of the 18 main cables supporting the platform is a bundle of 160 eight cm diameter wires with the bundle painted over and dry air is continuously blown through to prevent corrosion due to the humid tropical climate.
Each of the three towers is back-guyed to ground anchors with seven 8.25cm diameter steel bridge cables. There is one tower on the north side named T12, one in the SE corner named T4 and one in the SW corner named T8. The tower names appear to be based on position in relation to a clock face.
Another system of three pairs of cables runs from each corner of the platform to large concrete blocks under the reflector. They are attached to giant jacks which allow adjustment of the height of each corner with millimetre precision.
So the dish remained stationary and the reflector moved on a series of cables depending on the telescope positioning. Think like the Skycam at sporting events, like American Football.
Once radio waves from space hit the dish, they were collected by detectors in a dome above it.
The detectors could move along a cable line above the dish.
The antennas are very sensitive and highly complex radio receivers. They are contained in a bath of liquid helium to maintain a very low receiver temperate.
Low receiver temperate is important to minimize electron noise in the receivers, which allows for better detection of very weak radio signals.
On September 21, 2017, high winds associated with Hurricane Maria caused the 430 MHz line feed to break and fall onto the primary dish, damaging roughly 30 of the 38,000 aluminum panels. Most observations didn’t use the line feed, operations could continue. Restoring service levels would take more than the current operating budget.
On August 10, 2020, an auxiliary platform support cable separated from Tower 4, causing damage to the telescope, including a 30 m gash in the reflector dish. Damage included six to eight panels in the Gregorian dome, and to the platform used to access the dome. No one was reported to have been hurt by the partial collapse. The facility was closed as damage assessments were made.
The main support cables and towers had been designed with a safety factor of 2 to sustain twice the weight of the platform. When the Gregorian dome panels were added in 1997, it was believed that the cables would retain this safety factor, but due to uneven load distribution that would be almost impossible to guarantee.
Recent calculations and modelling of the structure found that the safety factor tower T4 had dropped to 1.67. But they believed the structure was still safe for the time being until repairs could be made.
There were also periods of time where the fans flowing air through the cables were not operating.
They found this in the 1980s and again in the 2010s that the molten zinc which affixes the cable to the socket mount at the tower was not complete, which allowed moisture to penetrate the wire bundle and increase corrosion.
Plans were made to replace all six auxiliary cables from tower 4 at a cost of $10.5 million USD
But before that could happen, a second cable broke on November 7th, 2020 and shattered part of the dish. The National Science Foundation announced that the telescope was in danger of collapse and the cables could not be safely repaired.
On November 19, 2020, after reviewing proposals to stabilize the telescope or reduce the weight of the platform, it was announced they would decommission the telescope.
On December 1, 2020, the final cable from tower T4 snapped, causing the platform to crash into the side of the dish.
Leading up to this, the cables supporting the towers had been failing at a rate of one or two a day.
When the platform collapsed, the backstay cables on the remaining towers, which had been adjusted to provide more support away from the dish, caused damage to those towers as well. Even causing minor structural damage to other buildings on the campus.
The cable that failed in November dated back to the observatory’s construction in 1963. The socket joint design did not establish the end of life capability.
There are videos of the collapse from the ground and air viewpoints. We will provide links to those in the show notes on our website www.failurology.ca.
The design factor of safety was significantly less than the minimum suggested and didn’t ensure structural redundancy in the event of a cable failure.
NASA Report Findings
NASA issued a report of findings in June of this year. It’s very long, so we summarized it for you. Before we get into their findings, we have a quote from the report that we think is important.
“The NASA/Aerospace team concludes that the most probable cause of the Aux M4N cable failure was a socket joint design with insufficient design criteria that did not explicitly consider socket constituent stress margins or time-dependent damage mechanisms. The socket attachment design was found to have an initially low structural margin, notably in the outer socket wires, which degraded primarily due to zinc creep effects that were activated by long-term sustained loading and exacerbated by cyclic loading. Additionally, a few wires showed evidence of hydrogen-assisted cracking (HAC) and wire surface defects that may have contributed to initial outer wire failures.” – From Arecibo Observatory Auxiliary M4N Socket Termination Failure Investigation
Wires were traced and labelled to indicate whether they had fractured, whether they had slipped from the socket and their location of termination
Of the 126 cable wires, 56 fractured within the socket, and 70 did not fracture, instead of pulling free of the socket joint.
Of the 56 that failed, five had surface defects running along their lengths. Two of those defects likely contributed to the failure.
The cable section that pulled free from the socket joint had 94 wires still encased together in zinc from the socket cavity, referred to as the cable/zinc slug. Meaning 32 wires were not in the socket cavity.
Examination of the wires from the failed cable showed that of the 94 wires in the cable/zinc slug, 26 were fractured wires that mated to the fractured wires in the socket and 68 were intact wires that did not fracture. Of the 32 remaining cable wires not accounted for in the cable/zinc slug, 30 wires were fractured and mated to the fractured wires in the socket and 2 were intact wires that did not fracture; all 32 were outer ring wires. In other words, 0 to 2 intact wires pulled free of the socket individually, and the other 30 to 32 outer ring wires came free of the cable/zinc slug due to insufficient surrounding zinc and/or from the forces after core pullout and subsequent impact.
Also as the wires and slug failed over time, the loads were redistributed to other wires. Several wires showed brittle failure which can occur under high-impact loading.
While some corrosion protective measures were put in service during the life of the socket (e.g., a mastic coating on the casting cap), these measures were put in place after corrosion had begun and were not adequately maintained to provide continued corrosion protection over the life of the socket. This resulted in pervasive quantities of corrosion product, particularly zinc oxide, throughout the socket along various identified moisture pathways.
In summary - failure of the outer wires occurred prior to total cable collapse. The outer wires were critical in maintaining the function of the socket joint and the outer wires were highly stressed with minimal structural margins of safety.
China has a similar telescope and stated they would start taking applications for international researchers to use their telescope in 2021
In late December 2020, the governor of Puerto Rico signed an executive order to remove debris and design a new observatory in the telescope place. The order also designated it a heritage site.
Plans are already being developed to replace the site with 1000 closely spaces 9m wide telescopes mounted on a flat plat over the Arecibo sinkhole. While the telescopes would be fixed, the plate could move to allow 500 times the field of view of the original.
So there you have it, a combination of small safety factors, questionable maintenance and unfortunate weather and timing led to the collapse of the Arecibo telescope before it could be safely repaired or decommissioned. But luckily no one was hurt, and who knows, they could end up replacing it with something way better that they wouldn’t have done if the telescope was still functional.
For photos, sources and an episode summary from this week’s episode head to Failurology.ca. If you’re enjoying what you’re hearing, please rate, review and subscribe to failurology, so more people can find it. If you want to chat with us, our Twitter handle is @failurology, you can email us at firstname.lastname@example.org, or you can connect with us on Linked In. Also, please check out our Patreon page for mini failure episodes. Check out the show notes for links to all of these.
Thanks, everyone for listening. And tune in to the next episode where we’ll talk about Apollo 1, the space program that eventually led to Americans walking on the moon. Bye everyone, talk soon!