With the launch date for Discovery mission STS 114 fast approaching, NASA still faces steep obstacles to returning the Space Shuttle to flight
Getting into orbit will be only one of the challenges when the Space Shuttle returns to flight, whether that is in March next year, as is planned, or later, as is likely. Once in orbit, the Shuttle will have to undergo a complex series of inspections and, if damage is detected, a difficult sequence of repairs. And then there will be re-entry.
With the scheduled launch of the Shuttle Discovery on mission STS 114 barely nine months away, it is no surprise that just two of the Columbia Accident Investigation Board's (CAIB) 15 return-to-flight (RTF) recommendations have emerged as the greatest challenges. Paraphrased, the two recommendations require NASA to eliminate all external tank (ET) debris shedding and develop the capability to inspect and repair damage to the thermal protection system (TPS) in orbit.
While much progress has been made towards meeting these and other RTF requirements, preventing debris shedding and developing inspection and repair techniques remain on the critical path to a return to flight between 6 March and 18 April next year. "We have made significant progress across the board in all the items that we think are required to return to flight," says Wayne Hale, deputy Shuttle programme manager. But former Shuttle commander Richard Covey, co-chairman of the independent task group charged with assessing the return-to-flight effort, says "there is much work to be done before NASA is truly ready for RTF".
Columbia disintegrated on re-entering the atmosphere in 1 February 2003, at the end of 17-day science mission STS 107, killing all seven astronauts on board. The cause was a breach in the reinforced carbon-carbon (RCC) leading edge that allowed superheated air to enter the wing and melt the structure. The breach was caused by insulation foam that broke away from the external tank, 81s after lift-off on 17 January, and struck the orbiter's left leading edge at RCC panel 8
The chunk of foam, estimated to be 530-690mm (21-27in) long and 300-450mm wide, separated from the ET's left bipod ramp, an aerodynamic fairing over one of two fittings for struts attaching the orbiter to the external tank, and hit Columbia at around 850km/h (530mph). During the accident investigation, ground tests proved that a 0.76kg (1.67lb) piece of foam impacting at 236m/s (775ft/s) could punch a 400mm hole in RCC panel 8.
NASA has concluded that the root cause of the foam loss was the migration of gaseous or liquid nitrogen, formed in the region between the liquid-hydrogen and liquid-oxygen tanks as the ET is filled, into voids in the manually sprayed foam. After lift-off, warmed by airflow over the tank, nitrogen trapped in these voids expanded explosively, throwing off pieces of foam, including the suitcase-sized chunk of bipod ramp that struck Columbia.
Acknowledging it is probably impossible to eliminate foam shedding totally, NASA is making a "best effort" attack on the problem by identifying the minimum debris size that can cause critical damage to the orbiter TPS. Earlier this year, the latest analysis of how the complex airflow around the Shuttle would transport the foam forced the agency to expand the critical debris zone and significantly reduce the allowable debris mass.
The critical zone has been increased from ±67.5° to greater than ±80° from the top of the ET, which faces the underside of the orbiter, and the allowable mass has been reduced from 0.09kg in this region to just 0.018kg.
NASA has a three-phase approach to eliminating the potential for debris loss, only the first stage of which will be completed before return to flight. While Phase 2 calls for automating critical manual spray processes, Phase 3 potentially involves redesign of the ET to eliminate the foam, although this measure may not make sense if the Shuttle is to be retired as planned once assembly of the International Space Station (ISS) is complete.
Under Phase 1, the agency is tackling known sources of critical debris, including the bipod ramps, liquid oxygen feed line bellows, protuberance air load (PAL) ramps and LH/intertank flange. This work has to be completed by mid-September if the redesigned ET-120 tank for the STS 114 RTF mission is to be shipped to Kennedy Space Center, Florida in time for a March launch.
The forward bipod fittings have been redesigned to eliminate the spray-on foam and redundant heaters incorporated to prevent ice formation. The critical design review was completed in April, with retrofit planned for July. The bellows that allow liquid-oxygen feed line motion, but also the potential for ice formation, are being redesigned with a "drip lip" added to the insulation to divert condensation and reduce ice formation. Validation is set for this month.
Redesign of the manually sprayed PAL ramps, which reduce aerodynamic loads on the ET's external cable trays and pressurisation lines, is planned for Phase 2. For Phase 1, NASA is revalidating the existing design and implementing non-destructive inspection (NDI) to increase confidence. Terahertz imaging and backscatter radiography are the NDI methods selected to detect any critical defects in the foam ramps.
Closing out the mechanically joined LH/intertank flange, with its complex geometry and history of foam loss, has emerged as the pacing activity on the tank for RTF. While studies to determine the critical size and transport mechanism for debris from this region continue, NASA has decided to strip and reapply foam on a larger area and elected to adopt an enhanced manual closeout procedure to eliminate voids and seal leak paths. This involves filling the intertank crevice with foam to reduce liquid nitrogen formation.
NASA has decided that the enhanced closeout process will be extended around the intertank flange ±112¡, to include thrust panels on either side of the ET to which with the Shuttle's solid rocket boosters (SRB) are attached. To meet the tank shipment deadline for a March launch, closeout has to be completed by the beginning of September, which NASA admits is a "significant challenge", but believes is achievable.
In its second interim report, published in mid-May, the Stafford-Covey Return to Flight Task Group concurred that three of the CAIB's 15 RTF recommendations had been fulfilled by: ensuring that at least two NASA employees attend all final close-outs and intertank hand spraying; implementing an inspection plan to determine the structural integrity of all RCC components; and reaching agreement with the US National Geospatial-Intelligence Agency to provide reconnaissance-satellite imaging of each Shuttle flight while in orbit.
"Substantial progress" has been made on almost all the 12 remaining RTF recommendations, says the report, and the Stafford-Covey task force expects "several more" will have been met by the time of its next interim report, expected around September/October. Among these is likely to be the requirement to improve ground-based imaging of the Shuttle from lift-off to at least SRB separation.
Upgrading the ground-based cameras is one of a series of changes intended to help detect and locate any debris damage that might occur on ascent. The lack of high-speed, high-resolution cameras contributed to the Columbia disaster and hampered the accident investigation. NASA's solution is to add cameras and upgrade to high-definition television. Airborne imaging of the Shuttle's ascent from a NASA Martin WB-57F high-altitude research aircraft is under consideration.
Additionally, to provide engineering-quality images of the external tank after it separates, NASA is replacing the 35mm- film camera in the orbiter umbilical well with a high-resolution digital camera. NASA will also provide the crew with handheld digital still cameras with telephoto lenses. Images from these cameras, down linked once the Shuttle is in orbit, will allow engineers to look for missing foam on the tank.
NASA also plans to install digital cameras on the ET liquid oxygen feedline fairing to view the bipod area and underside of the orbiter, and on the forward skirt of each SRB to view the intertank area. High-resolution imagery from these cameras will be downlinked live to mission control. Plans call for the mounting of additional cameras on the ET and SRBs, to improve coverage of the orbiter's wing leading edges, as soon as possible. They will not be ready for STS 114.
Although not an RTF requirement, or constraint, NASA plans to install impact sensors in the wing leading edges. There will be 92 sensors per side, each a three-axis accelerometer able to sense high-g events that might indicate a high-energy impact. The sensors will not be able to determine whether damage has been inflicted, or how much, but could help focus inspection efforts once in orbit. Initially at least, all of the leading edge will have to be inspected even if no impact was detected.
NASA's greatest challenge in returning the Shuttle to flight is in meeting one key CAIB recommendation: to develop the capability to inspect and repair the orbiter's thermal-protection system, both tile and RCC, in space. Already the agency has acknowledged that it will be unable to develop such a capability for flights that do not involve docking with the ISS, controversially cancelling a mission to service the Hubble Space Telescope because it could not meet the CAIB's requirements. Instead, NASA is focusing on inspection and repair schemes that use ISS resources, at least for RTF.
As a first step, NASA has to decide what constitutes critical damage, as this drives the inspection resolution and repair techniques required. The goal is to define damage thresholds for all TPS zones below which no repair is required before re-entry. Detailed criteria, which include the extent and depth of the damage and the entry heating expected, are still evolving, but NASA has established preliminary inspection thresholds, including minimum crack-length resolutions as small as 6.5mm for the underside of the wing leading-edge RCC panels (see diagram below).
NASA's decision to develop a sensor-equipped boom to extend the reach of the Shuttle's robot arm and allow in-orbit inspection of the orbiter's nose, wing leading edges and underside has emerged as the pacing item for return to flight. To be supplied by Canada's MD Robotics, builder of the Shuttle and ISS remote manipulator systems (RMS), the 15m-long orbital boom sensor system (OBSS) will double the length of the robot arm, and its development is literally the long pole in the RTF tent.
According to the Stafford-Covey task group, "the OBSS schedule is very aggressive and has no slack time reserve". Stabilising and integrating images from the boom's television and laser sensors, and detecting and measuring damage with low contrast and in dark areas, are challenges highlighted in the group's latest report. NASA's Hale admits the boom has been problematic and agrees the OBSS is on the critical path to RTF, but he believes the system is beginning to come together.
Using the OBSS, NASA will avoid the need to dedicate the first days of a Shuttle mission to extravehicular activity (EVA) to verify there is no critical damage. Requiring spacewalking astronauts to inspect the entire orbiter could compromise the mission, which will involve assembly and resupply of the ISS. "The boom has become important," says Covey. "If it is not ready to fly, EVA may be an option for the first flight, but it will not work long term."
Weighing around 400kg, the inspection boom will be stored on the starboard sill of the orbiter's payload bay until required on day two of the mission. For RTF, the boom-tip will be equipped with existing sensors: an intensified television camera and laser dynamic range imager mounted on a pan/tilt unit. The laser sensor will provide a three-dimensional image, allowing the depth of any TPS damage to be measured.
On day three of the RTF mission, on current plans, Discovery will approach the ISS from directly below and, at a safe distance of around 180m, perform a "pirouette" pitch-up manoeuvre through 360¡, at 1¡/s, to allow the ISS crew to photograph the orbiter using digital still cameras. Discovery will then dock with the ISS, where further inspection of any suspect areas will be conducted using the sensor boom and both the Shuttle and ISS robot arms. If an even closer inspection is needed, the boom will be able to support an astronaut on EVA.
The CAIB's recommendation requires NASA to develop the means to repair both tiles and RCC panels before return to flight. The tile repair scheme is well-defined, but RCC repair tools are still in the early stages of development - although promising progress has been made. Demonstrations of both tile and RCC repairs are planned for STS 114, but it is likely that, if real damage is sustained on the RTF mission, only the tiles could be repaired.
Tile repairs will use a silicone-based cure-in-place ablator that ground tests have shown adheres and cures in vacuum and survives the heat of re-entry, charring but not swelling enough to disturb the airflow over the orbiter. NASA is developing an applicator that will contain, mix and extrude the two-part material. The applicator - complete with vacuum-sealed reservoirs, pneumatic drive system, static mixer, hose and gun - will attach the astronaut's life-support backpack. The astronaut will also have a set of foam brushes, trowels and stamps for cleaning the damaged area, scraping away excess ablator, and smoothing the repair.
Three complementary repair methods for the RCC leading edges are being evaluated. Cracks and small holes would be filled with a putty-like material based on silicon carbide, developed by NASA's Glenn Research Center and called Graber. Holes 15-100mm in diameter would be covered by a plug of RCC held in place by a molybdenum bolt. Catastrophic damage would be repaired by wrapping the affected area with a rigid sheet of RCC, shaped to cover the most vulnerable panels.
To enable TPS repairs while docked at the ISS, the Shuttle will have to be turned to a belly-up position that provides access to the repair site using the ISS's robot arm. This "flip around" operation involves grappling the ISS with the Shuttle RMS, undocking the orbiter then using the arm to rotate it into position.
After the repair, the Shuttle RMS would rotate the orbiter back to its docking positions. The complex ISS-based inspection and repair procedures are already being practised in simulators.
The ISS-based repair procedure will work until the Japanese Experiment Module is attached, scheduled for the 15th mission after flights resume, blocking access to the grapple fixture. New inspection and repair techniques will have to be developed by then, admits NASA. These could include development of the Mini-AERCam, or autonomous extravehicular robotic camera, a small free-flying sensor platform.
If on-orbit repairs fail, or damage is too great, NASA plans to use the ISS as a "safe haven" for the Shuttle crew until it can launch a rescue mission with another Shuttle. For the STS 114 return-to-flight mission, the agency plans to have Atlantis ready to launch within 90 days, with the potential rescue mission designated STS 300. Use of the ISS as a safe haven "is becoming increasingly important in NASA's decision-making for return to flight", says the latest Stafford-Covey report, and the group plans to assess the capability as if it were a CAIB RTF recommendation. With no signs that NASA has begun the planning required to use the ISS as a safe haven, this could yet emerge as an obstacle to RTF.
Return to flight
Planned in the mid-1990s as a routine ISS resupply mission, STS 114/Discovery has taken on a distinctly different and critically important character. The Shuttle will still carry the Italian-built Rafaello multipurpose logistics module to ISS, and two of three EVAs planned involve replacement of a failed control moment gyro on the ISS and installation of an external stowage platform on its Quest airlock, but the once-planned ISS crew switch has been dropped.
Instead, attention will focus on the first EVA, planned for day five, when mission specialists Stephen Robinson and Soichi Noguchi will test TPS repair techniques in Discovery's payload bay, using the robot arm and a specially designed fixture. A second RTF test mission, STS 121/Atlantis, has been scheduled for launch no earlier than 5 May next year. Only after two successful flights - and two safe re-entries - will the Shuttle be allowed resume its remaining tasks: to complete assembly of the ISS and then retire.
GRAHAM WARWICK / WASHINGTON DC & TIM FURNISS / LONDON
Source: Flight International