No abstract

INTRODUCTIONSpace Shuttle Discovery's last mission, STS-133, was scheduled to launch on November 5, 2010. Just hours before liftoff, a hydrogen leak at an umbilical connection scrubbed the launch attempt. After the scrub, further inspection revealed a large crack in the foam insulation covering the External Tank, ET-137. Video replay of the launch attempt confirmed the crack first appeared as cryogenic propellants were being loaded into the ET. When the cracked foam was removed, technicians found the underlying stringer had two 9-inch-long cracks. Further inspection revealed a total of 5 of the 108 ET stringers had cracked. NASA and Lockheed Martin immediately launched an aggressive campaign to understand the cracks and repair the stringers in ET-137, targeting February 2011 as the new launch date for STS-133. Responsibilities for the various aspects of the investigation were widely distributed among NASA centers and organizations. This paper will focus on lab testing at Marshall Space Flight Center (MSFC) in Huntsville, Alabama that was intended to replicate the stringer failure and gauge the effect of proposed countermeasures. BACKGROUNDThe ET's function is to carry the cryogenic propellants that feed the three Space Shuttle main engines. The ET also receives and distributes thrust from the Solid Rocket Boosters (SRBs). On-going ET production allowed evolution of the design, and ET-137 represented the latest "Super-Lightweight" configuration, which made extensive use of Aluminum-Lithium alloys. Figure 1 shows the layout of the Super-Lightweight ET. The Intertank connects the liquid hydrogen (LH2) tank on the aft end and the liquid oxygen (LOX) tank on the forward end. Unlike the propellant tanks, the Intertank is not stiffened by internal pressure, so stringers are included to add stiffness and strength. These are the same stringers that cracked on ET-137 and are the focus of this investigation. The Intertank is constructed of eight panels, six of which are skin-stringer construction with a typical cross-section as shown in Figure 2. The stringers are mechanically attached to the skin with rivets along most of their length and with specialty fasteners, such as GP Lockbolts and Hi-Loks, at the forward and aft ends. When the propellants are loaded, both ends of the Intertank approach cryogenic temperature and contract inward. The center of the Intertank remains at warmer temperature and resists contraction, causing considerable bending of the stringers, especially at the ends. Figure 3 shows cracks on the LOX end of a stringer that are consistent with this loading scenario, appearing as if the stringer end pulled away from the skin, fracturing the stringer foot along the fastener line. Fractography analysis indicated the ET-137 fractures were static, initiating on the bottom of the stringer foot (the tensile side of bending) between the 2 nd and 3 rd fasteners from the end. The video footage of the launch attempt indicated that the fracture occurred precisely when the LOX level should have reached the Intertank/...

No abstract

No abstract

The sealing capability of the Ares-I Upper Stage liquid oxygen tank-to-sump joint is assessed by analyzing the deflections of the joint components. Analyses are performed using three-dimensional symmetric wedge finite element models and the ABAQUS commercial finite element software. For the pressure loads and feedline interface loads, the analyses employ a mixed factor of safety approach to comply with the Constellation Program factor of safety requirements. Naflex pressure-assisted seals are considered first because they have been used successfully in similar seal joints in the Space Shuttle External Tank. For the baseline sump seal joint configuration with a Naflex seal, the predicted joint opening greatly exceeds the seal design specification. Three redesign options of the joint that maintain the use of a Naflex seal are studied. The joint openings for the redesigned seal joints show improvement over the baseline configuration; however, these joint openings still exceed the seal design specification. RACO pressure-assisted seals are considered next because they are known to also be used on the Space Shuttle External Tank, and the joint opening allowable is much larger than the specification for the Naflex seals. The finite element models for the RACO seal analyses are created by modifying the models that were used for the Naflex seal analyses. The analyses show that the RACO seal may provide sufficient sealing capability for the sump seal joint. The results provide reasonable data to recommend the design change and plan a testing program to determine the capability of RACO seals in the Ares-I Upper Stage liquid oxygen tank sump seal joint.