Once the SRB ignition is complete and combustion pressure has reached its full operating value, there is a greater effect on the walls than on the joint because the joint is more rigid to hoop stress. Thus, the joint rotates and the gap expands. The last thing you want.
https://www.ntsb.gov/investigations/AccidentReports/Reports/AAR0201.pdf
Let me reiterate, aircraft must be maintained with care and at all cost.
Alaska Airlines expanded rapidly in the years before this accident. With the goal of becoming more profitable, as they became bigger and busier the pressures to keep their planes on schedule put increasing stress on their maintenance facilities. And this took its toll. The Federal Aviation Administration seemed to know this, but was nowhere effective in preventing the tragic chain of events.
So N963AS began its fateful path in a C-check years before falling to the ocean. Its maintainers found a jackscrew that needed to be pulled, but no spare was found and, as the part was arguably acceptable, they pushed the plane back into service, with no watch list, no trailers, or orders to keep track of its condition. There were no specific procedures to do so, and no one thought enough to ask for one.
The aircraft was arguably, that is technically, legal, and it was probably safe, if it was carefully greased. It was not, we know that without question. And the lengthened inspection intervals were such that it was not to be looked at again, until it was in our laboratory.
When it finally failed, ground support from Alaska Airlines seemed to encourage the crew to proceed with a broken plane on to their scheduled destination, for reasons perhaps of convenience both to passengers and maintenance -- we won't really ever know.
But the impression is inescapable. An aircraft that had been hustled out the door three years earlier for the convenience of scheduling was now encouraged to keep to its appointed routing. It is less coincidence than culture.
By far the best description of Hubble's spherical aberration you will see.
https://iopscience.iop.org/book/978-0-7503-2038-2/chapter/bk978-0-7503-2038-2ch1
The conference room in Goddard's Building 3 was packed that afternoon as the lead engineers gave technical status reports about the various spacecraft systems for which they were responsible—thermal, electrical, pointing control, computers, science instruments, and telescope optics. The Optics Lead Engineer, Charlie Jones from Marshall, reported a continued lack of success in attempts to find an alignment and focus position of the telescope optics that would produce sharp stellar images. At that point, Chris Burrows, an astronomer and optics expert from the STScI, stood and said, with an edge of anger in his voice, "You have about a half wave of spherical aberration, and there's nothing you can do about it!"
Behind the scenes Chris Burrows, and independently Jon Holzman and Sandy Faber from the WFPC Team, had been analyzing the Hubble images. They had correctly deduced the nature of the problem with the telescope's optics, namely spherical aberration (Burrows et al. 1991). Up to that time, the Marshall engineers and the telescope's manufacturer, Perkin-Elmer Corporation in Danbury, Connecticut had ignored the astronomers' conclusions and denied that anything was wrong with the telescope. That's why Chris was angry. He had been frustrated for some time trying to convince the telescope makers that "the emperor had no clothes," that the telescope's primary optics were defective. They didn't believe it. They didn't want to believe it. But now Chris, Sandy, and Jon felt that their evidence was ironclad. The room resonated with a stunned silence.
The designers of Hubble had anticipated that, in going from 1g on the Earth's surface to the microgravity of low Earth orbit, the telescope's 2.4 m primary mirror might deform slightly so as to produce a variety of "higher order" aberrations or distortions of the focused image. For this reason, the telescope's design included 24 pads situated behind the primary mirror, that could be actuated in various combinations so as to apply gentle pressure to the back of the mirror to adjust its shape and remove the aberrations. The one aberration that could not be corrected in this way was the simplest of optical aberrations—spherical aberration.
Every amateur astronomer who grinds his or her own mirrors by hand tests for spherical aberration and removes it in the grinding process. But one of the world's most sophisticated telescope makers, Perkin-Elmer Corporation, who reputedly had produced a number of similar large mirrors for Earth-observing spy satellites for the U.S. Department of Defense, had failed to detect this most basic of aberrations in the primary mirror they had built for Hubble.
Almost a decade later Lew Allen and his Board uncovered what had happened during their investigation on the scene at Perkin-Elmer. They demonstrated that the 1.3 mm error in the spacing between the lens and the lower mirror in the reflective null corrector, and the corresponding polishing error in the shape of the Hubble telescope's primary mirror, precisely explained the amount of spherical aberration observed in the camera images Hubble was sending back from space.
Ironically, two other optical test devices clearly showed the mirror's defect in data they provided at the time, independent of the reflective null corrector. But both Perkin-Elmer and NASA chose to ignore this, having become convinced that those other two instruments were not sufficiently accurate to provide credible data at the high level of precision required. They thought only the reflective null corrector could do that. The Allen Board demonstrated that this was not true. Both of the other devices were accurate enough to reveal a major, gross error of the magnitude that had been built into the Hubble mirror.
The investigations revealed myriad failures of management oversight, technical processes, and quality control on the part of both Perkin-Elmer and the NASA Hubble Project. There were a lot of excuses. Perkin-Elmer was badly behind schedule and over budget in manufacturing both the telescope and the Hubble fine guidance sensors for which they were also responsible. So there was much pressure to work quickly and perhaps to cut corners. The division of Perkin-Elmer that had the job of grinding the mirrors and conducting the optical testing operated in a closed-door environment. According the report of the Allen Board, neither people from NASA nor from other divisions of Perkin-Elmer had easy access to the Optical Operations Division for purposes of oversight.
Although the Department of Defense denied any responsibility for this state of affairs, the review team speculated that a culture of secrecy prevailed within that group because it reputedly had also produced optical systems for classified satellite payloads in the same plant. Finally, the small cadre of personnel the Hubble Project at Marshall assigned to monitor the work at Perkin-Elmer really did not have the experience or expertise needed in the area of large optical telescope systems.
