alternatehistory.com

V: Number One On the Runway

As the Space Shuttle program entered into mid-1979, it had found itself in trouble. An SSME failure had hit the program first, by delaying the first flight into 1980 and in addition further delaying the minimum amount of seconds of operation before flight. Then came the news over the tiles. The fact that NASA was now having to deal with figuring out a way to make sure that the tiles wouldn’t come off during the first flight was a ‘gut-punch’ to the space agency. NASA would rapidly begin a search for a solution to the tile program, so that the tiles were bonded to the Orbiter and would not come off during flight. By August, NASA had managed to figure out a solution dealing with the way the tiles were bonded to the felt pads they were connected to. An analysis indicated that each of the individual components making up the connection of the thermal protection system (TPS) to the Orbiter were fine (the tile, the SIP, and two layers of adhesive) but when they were combined together it was shown that they lost about half of the combined strength of it all. The reason for the loss in strength had been traced to ‘stiff spots’ present in the felt pads that the tiles were connected to.

The solution for this would be found in a ‘densification’ between the tiles and the felt pads. The densification process would fill the void between the fibers next to the felt pad with a special slurry mixtures that would remove the ‘stiff spots’ present. The densification process however was quite arduous, requiring both the removal and re-addition of the waterproofing on the tiles, air-drying the tiles for twenty-four hours after the slurry mixture was ‘painted’ on the back of them, and then requiring to ‘bake’ the tiles in an oven. Nonetheless, the slurry mixture would act as a new layer on the bottom of the tile acting as a ‘plate’ to eliminate the stiff spots on the felt pads. However, there was a downside to the tile densification process. It would require the removal of all the tiles so that they could be densified and then reinstalled back on the Orbiter.

Because of the amount of time that would need to be taken to both remove and then reinstall all the tiles, a challenge emerged in order to save as many of the installed tiles as possible while also ensuring that those tiles that remained installed would have a structural margin safe enough for flight. The solution developed for this would be known as the ‘tile proof test’. The tile proof test involving applying a load to the installed tile equivalent to 125% of the maximum flight stress that would be experienced by that tile at the highest point of stress. While the process would help to save thousands of installed tiles, it would also reveal those tiles that had an inadequate flight strength and would need to be replaced. In addition to the tile proof test, there were other methods designed to help strengthen the tiles while they were still attached. This entailed the use of something known as ‘gap fillers’ to help fill the space between the tiles to prevent them from rotating by shock waves of air during reentry and coming off. For the ‘thick’ tiles on the underside of the Orbiter, they could be added as needed, but for the small thin tiles where it was not possible, the gap fillers would be directly bonded to the tiles.

At the same time as work on replacing the tiles were occurring, the Air Force Flight Dynamics Laboratory were performing a series of tests to review the Orbiter for potential heating concerns. The tests would reveal that the OMS pod were going to be deflecting much more heat than had been expected, and because of the existing tiles being relatively weak under existing loads, there were concerns that it could fracture and separate from the Orbiter during reentry. Because of the tiles having already been installed to the OMS pods, a solution was developed in order to be able to prevent their fracture and separation from reentry. The 8-by-8 inch tile would be ‘diced’ into nine equal parts while it was still attached to the OMS pod and to ensure that it would neither damage the tile nor the structure of the OMS pod underneath it. [1]

As the work on the tiles progressed, significant efforts were also underway for the SSMEs. While it had arguably been completed in terms of the ‘developmental’ testing following the failure of a test engine in mid-January, it was now the goal of the program in making sure that the engine was capable of flight. John A. Yardley, the Associate Administrator for Space Flight had set the directive of achieving 65,000 seconds of single-engine test operation for the SSME prior to flight. By the end of 1978, only 34,118 seconds had been completed of single engine testing, and prior to the tile issue having emerged it had seemed as if they would barely get the needed 65,000 seconds of testing prior to flight. In addition to the requirement for the 65,000 seconds of single engine testing, there was also that of directive to ensure that the engines could run for a full flight duration without failure, alongside the testing for the variety of abort modes if needed.

As the testings resumed on the SSME, issues would again sprout on the SSME and the need to fix those new issues that had sprouted up. The issues would sprout up most often on the MPTA where you had three of the SSMEs to test for flight configuration. For example during one test, a failure from one engine would show a need of redesign on a main fuel valve while a failure from another engine during that same test while throttling down would highlight the issue of a hydrogen tube in a nozzle breaking (this was principally to help cool the engine). While work would commence on redesigning the hydrogen lines (by stiffening them) to prevent another failure like that happening, it would be uncovered that improper welding had contributed to the failure of the nozzle (with a series of fixes to those nozzles that had seen the improper welding job). But the issues and failures with the SSMEs would decrease further and further and by February, 1980, the SSME had managed to achieve a total of the 65,000 seconds of testing, passing the requirements set by Associate Administrator Yardley. But the Space Shuttle was still not ready for flight as work continued to progress at Kennedy for the replacement of tiles for Columbia.

Throughout the rest of 1980 work continued on preparing the Shuttle for her eventual first flight as the launch date continued to slip throughout 1980 and then into 1981. While progress was being made on the tiles, it was felt at times that progress was going nowhere as it was being found that the same amount of tiles were being installed as were being removed (if not taking off more than being installed by those newly-installed tiles having failed the tile proof test). But despite that, the steady progress/march towards flight continued as the amount of tiles that had to be installed continued to chart downwards. Both the External Tank and the Solid Rocket Boosters had finished their testing, and the SSME, one of the most troublesome pieces of the program had found itself running smoothly, performing multiple full duration burns [2] along with performing the kind of burns necessary in the event of an engine-out scenario and the associated abort mode [3].

By the start of November, the work had been finished and Columbia was rolled out to the Vehicle Assembly Building to be stacked and mated to the Solid Rocket Boosters and External Tank. After nine years of development and seeming setbacks, the Space Shuttle’s first flight, STS-1, was set for March 10th. The Space Shuttle was ready to go.


[1] The section on the TPS and the tile replacement is explicitly referenced from Space Shuttle: The History of the First 100 Missions, Page 239-240, in part considering the specific details on the densification process and other associated details to help with dealing the issues encountered.

[2] Prior to the launch of STS-1, more than one hundred thousand seconds of testing of the SSME would be completed.

[3] In the event of an engine-out scenario from launch to separation of the ET, a series of abort modes would be available. No abort could be made until after the SRBs had finished burning and could be separated from the vehicle. Starting at T+2:10, the 'Return to Launch Site (RTLS)' abort mode could be selected. The 'RTLS' abort mode would be available through the first four minutes of flight, and then as originally proposed would be followed by that of the 'Abort Once Around (AOA)' (in which it would abort and then land at Edwards after a single orbit) or 'Abort to Orbit (ATO)' (having arrived at a lower orbit than expected). In between the RTLS abort mode and the ATO abort mode, the Shuttle would be expected to be ditched. In the lead-up to STS-1, Dick Truly and Joe Engle would develop a new abort mode to prevent such a ditching known as the 'Trans Atlantic Landing (TAL)' abort mode. This abort mode would be available starting at a later point then when the RTLS abort mode could be initiated, but would extend up to the AOA/ATO abort mode selection along with being safer for the Space Shuttle (in terms of the flight path and stresses on the Orbiter). However, because of the development of the TAL abort mode, it would not be available in the Shuttle flight software through the orbital test flights and would be expected to be hand-flown by the flight crews (it would be available starting in the first operational mission).