Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
NATIONAL RESEARCH COUNCIL COMMISSION ON ENGINEERING AND TECHNICAL SYSTEMS 2101 Constitution Avenue Washington, D. C. 20418 COMMITTEE ON NASA SCIENTIFIC AND TECHNOLOGICAL PROGRAM REVIEWS Panel on Redesign of Space Shuttle Solid Rocket Booster The Honorable James C. Fletcher Administrator National Aeronautics & Space Administration 400 Maryland Avenue, S.W., Room 7137 Washington, DC 20546 Dear Jim: January 15, 1987 I am pleased to submit herewith the third interim report of the National Research Council's Panel for the Technical Evaluation of the Redesign of the Space Shuttle Solid Rocket Booster. The redesign effort is currently in transition from a period devoted to creating the so-called baseline design using analytical techniques to one concentrating on testing the concepts and hardware. The purpose of this letter is to critique the program at this stage. Background Since early October, the Panel has met formally three times. In addition, groups of our members have participated in a number of technical meetings with personnel from Marshall Space Flight Center and Morton Thiokol, Inc. Members have visited Latrobe Steel Company and the Ladish Company to review capabilities for producing steel cases for the motor and the U.S. Air Force Space Command to discuss methods of nondestruc- tive evaluation of insulation bonds, as well as other aspects of the redesign that might benefit from the Titan SRM program. We have also kept up to date on the progress of the contrac- tors who are developing concepts for a Block II motor. Current Status NASA's goal is to return a safe and reliable Shuttle to service at the earliest possible date. To meet the goal, the NASA team has chosen a baseline design and is in the process 15 The National Research Council is the principal operating agency of the National Academy of Sciences and the National Academy of Engineering to serve government and other organizahons
Letter to the Honorable James C. Fletcher 2 of organizing and implementing a testing program to verify that the design meets the requirements of the program. The redesign baseline is aimed at correcting the causes that contributed to the failure as well as other elements of the original design that showed signs of trouble in flight or that do not meet the new design criteria. Most decisions have been made as to which features need complete redesign, which need modification, and which can remain the same as before. Only those changes considered mandatory for the next flight are being made. Only if the verification program goes very well can NASA's schedule be met; care must be exercised in the ground-test program to assure that the design is adequately demonstrated. Morton Thiokol is well started on this program and should be strongly supported in its pursuit. As discussed in an earlier report of our Panel, however, the test program is success-oriented. By this, we mean that the schedule for the program reflects an assumption that each test will produce results that are expected and understood; there is little room in the schedule for modifying the design if this does not prove to be the case. Preliminary tests have already produced results that were not anticipated. For exam- ple, O-ring materials with better low temperature resilience were found to deteriorate when tested after extended exposure to the rust-inhibiting grease, with the result that the base- line design now uses the original fluoropolymer O-rings with heaters to maintain an appropriate temperature. It is prudent, therefore, to plan for contingencies, particularly for those elements of the design for which substantial uncertainties remain. If an aspect of the design is shown not to meet safety objectives, an alternative should be available to carry the program forward. In some areas the program has incorporated alternatives to the baseline design. Nonetheless, we are concerned that valuable time would be lost should it become necessary to turn to one or more of the alternatives as a consequence of something learned from the test program. Further, now that the NASA/Thiokol redesign team is nearing the completion of the selection of the base- line design, it should be able to increase its efforts on preparing for contingencies. We recommend, therefore, that NASA strengthen its contingency plans for incorporating alternatives into the program. Aspects of the design that warrant more contingency planning include the case field joint, the case-to-nozzle joint, and configurations of the insulation at both the case and case-to-nozzle joints. Materials for which contingencies 16
Letter to the Honorable James C. Fletcher should be in the program include the ablative materials in the nozzle, O-rings, and the rust inhibitor. Options currently in the program are limited. For example: o The Panel shill believes that contingency planning should include the design, fabrication, and testing of a case field joint that closes under pressuriza- tion. We understand that such a design can be obtained without changing the configuration of the existing forgings, but the first step in this process is to order the appropriate forgings, which has not been done. O Currently, there is no real alternative in the program for the baseline design of the case-to-nozzle joint, which incorporates a large number of new bolls and bold holes, hence complex stress patterns and potential leak paths. We consider the lack of an alternative to be serious since the joint is critical for safety, few tests of the final configuration are planned, and they occur late in the test program. We recommend that alternatives be established and l special emphasis be placed on early, meaningful tests of the design. a The baseline design of insulation at the field case joint incorporates an adhesive bond between opposite sides of the gap. Two variations of this unvented design are being studied, called the J-seal and U-seal. The backup design, a vented configuration, will be evaluated in subscale tests and a ful1- duration, fu11-scale test firing, but it is not being carried forward on a schedule that would permit its inclusion in the program without a delay. The base- line design of the case-to-nozzle joint also incorpo- rates an adhesive bond. NASA should anticipate that the details of the insulation design at both joints, whether unvented or vented, will continue to change and we recommend that the Agency provide for addi- tional insulation tooling to enhance the flexibility of the program. Significant changes in estimates of loads in the SRB and both of its interfaces with the External Tank have recently been reported to us. In light of these changes and the crack in an aft skirt during a recent structural test, we recommend that stress analyses of the booster and its parts be carefully reviewed and, as appropriate, calculated again using the best data and most up-to-date models. 17
Letter to the Honorable James C. Fletcher The Panel has additional technical comments and recommen- dations on the current design and testing program, which are presented later in this report. Perspective on the Future .,, i . Not all the changes that the engineering team would like to make in the solid rocket motor will be made before the next launch. By directive from NASA headquarters, the redesign effort is focussing on identifying those changes essential for adequate safety or reliability for the next flight. The purpose is to avoid a dilution of effort on the mandatory changes, which could delay the Shuttle's return to service. We concur with this approach in principle. Beyond the changes deemed mandatory for the next flight, however, additional changes may also be needed further to enhance its safety and improve its reliability, performance, and cost effectiveness. The solid rocket booster, no less than other technological products, should continually be improved by changes driven by the results of tests and data gathered during use. It is our impression that the ongoing technological program that supported improvements in the reliability of the Shuttle solid rocket motor before the accident was limited compared with that for improving performance. We recommend, therefore, that NASA establish and maintain a directed program for continued evaluation and improvement in reliability after flights are resumed, including all of the following: use of flight instrumentation for diagnostics throughout the flight program; continuing evaluation of flight results and recovered hardware; continuing ground testing to understand and, as necessary, improve the hardware promptly; and preplanned block changes to the flight hardware to provide opportunities for incorporating desired improvements. In the long term completely new designs, possibly radi- ca1ly different from the current one, may be preferred. The nation's manned space flight program cannot afford to continue to rely on older technology indefinitely into the future. Consequently, we believe that NASA should vigorously pursue a program to develop the next and future generations of motors for use in the space transportation system. The studies of alternative concepts for the SRM that NASA commissioned in 1986 is a commendable step in this direction. The program outlined above--including near-term changes to meet immediate goals, intermediate-term changes for product improvement, and long-term changes to incorporate substan- 18
Letter to the Honorable James C. Fletcher 5 tially improved technology--can only be accomplished with the support of research and development. Not all of the required R&D can be accomplished within the current recovery program, so it may be necessary for NASA to include appropriate activ- ities elsewhere within its programs. Additional Conclusions and Recommendations on Technical Issues The Case Joint. We are concerned that the baseline design may not readily allow for reworking or inspection of the hidden surfaces of the capture feature. We conclude that this problem requires special attention since it could be suffi- ciently serious to prohibit reuse of the case segments if the finishes and absence of corrosion cannot be satisfactorily verified. The baseline design now specifies fluoropolymer O-rings lubricated with HD-2 grease, as used in all the previous flights, with heaters to maintain the O-rings at the tempera- ture (75°F) required to assure a satisfactory seal. The Panel is concerned with this development because of added complexities and potential reliability problems associated with the heaters. For example, overheating the joints could adversely affect adhesive bonds, which are invariably weakened at elevated temperatures, or the properties of components, such as O-rings or grease, in unpredictable ways. It seems likely that an alternative grease can be found or formulated that will both have acceptable corrosion protective properties and not interact adversely with the silicone or nitrile O-ring materials. We recommend that a priority effort be aggres- sively pursued to improve understanding of the properties of these materials and to develop a better combination of O-rings and grease for use as soon as possible. The redesign team has considered two concepts for the design of insulation between adjacent segments, vented and unvented, and has selected the latter as the baseline approach. In this concept, a properly functioning insulation would shield the metal joint from combustion gases. Should a leak occur in the interface, however, the joint has been designed to inhibit further propagation of the high-pressure gases through the O-ring seals. The small joint rotation anticipated with the capture feature, interference metal fit between the clevis leg and capture feature, and three O-rings are configured to provide redundant protection in sealing against various upstream leak scenarios. Because the unvented insulation design, if successful, will shield the O-rings from exposure to pressure on hot gases, tests with imposed flaws are particularly important for determining the adequacy of the design of the seals. 19
Letter to the Honorable James C. Fletcher 6 In response to the requirement that pressure seals be verified, the program envisions testing the O-rings after assembly by applying pressure through ports. Applying an inappropriate pressure could damage the insulation or bond line between case segments. We remain concerned with this issue and expect to pursue it further as verification procedures are developed. The Nozzle. The Panel currently perceives that uncertainties in the design and hence risk to the program are now greater in the area of the case-to-nozz~e joint and for the other aspects of the nozzle than in the case field joints. Unlike many of the issues associated with the field joints that can be resolved to a reasonable level of confidence with subscale or short duration tests, several of the critical questions on the nozzle design can only be really answered with full scale, full duration testing. Mating and sealing the segments of insulation at the case-to-nozzle joint is a very different proposition from that at the case field joint. During assembly of the case-to- nozzle joint, the insulation on the fixed nozzle housing must slide for several inches over the insulation on the aft dome, complicating design of the insulation and adhesive for an unvented design. Here the method of sealing of the mating segments of insulation is still open to question. We recommend continued model~ing and subscale testing of alternatives that insure that only cooled, low momentum gas can reach the O-rings in the event of a leak in the seal between mating segments of insulation at the joint. The general requirement for pressure seals to be redundant and verifiable is being applied in the current baseline design to all of the internal nozzle seals. This requires the addi- tion of a second O-ring in many cases and the addition of ports to verify and seat the seals. The second O-rings are intended to enhance reliability, but overall reliability may not be increased because of the penetrations of the nozzle added for the purpose of verifying the O-rings. Tests of the nozzle ablative materials will not be performed until late in the verification program. Marginal or unsatisfactory performance in these tests could severely affect the schedule for the next launch. There are no backups in the program for design of the Jayup of ablative materials in the nozzle. Given the empirical nature of the technology and its unpredicted behavior in past flights, we recommend that, in addition to the planned full scale, full duration tests of the nozzle, the redesign team undertake a series of subscale, long duration firings to evaluate the conceptual basis and materials technology for the redesign of the ablatives. 20
Letter to the Honorable James C. Fletcher 7 The Test and Verif ication Program . As indicated in an earl ier report, we regard the proof of the redesign to depend on the results of the program to test it and verify that it meets the requirements of the system. The test and verif ication program is just getting underway and is shill evolving. Some prelimi- nary tests have been conducted, but f irings with new case hardware are not expected before March, 19 87 . The test program appears to be somewhat constrained by the abil ity of the support organizations to provide test articles, instrumentation, etc. We intend to follow the development and progress of the test program carefully. Meanwhile, we make the following observations and recommendations. While we continue to see progress in the definition of the total test program, it is difficult to relate the objectives of each of the numerous tests to specific design requirements or to develop a sense of the coherence of the test program in terms of its support of the development effort. Clear plans that relate each test to specific design requirements or to the development of specific components or subsystems would help to assure coherence of the total effort. For NASA to accept a design as safe and reliable, we believe that the Agency should be convinced that it under- stands how the design will behave throughout the range of conditions that the SRB is intended to experience. Recog- nizing that it may not be practical to test the design throughout the range of conditions, NASA plans to verify some aspects of the design by analysis rather than by test. Verification by analysis relies on the use of analytical and computational models, so to incorporate this technique requires that the models be validated by experiment. We recommend that tests within the current program also be instrumented specifics ~ y to validate the models. The verif ication program is aimed at determining if the des ign meets requirements, but it should al so have the value of enhancing conf idence that the redesign team understands how the design works. The Presidential Commission concluded that ne ither NASA nor its contractors understood the previous des ign very wet ~ . To assure better understanding, we recommend that before each maj or test in the verif ication and qualification program the redesign team make expel icit analyti- cal predictions of the performance of the test article; the results of each test should be examined in light of the predictions and any discrepancies full y explained. Criteria for successful verif ication in each test should also be estabI ished a priori to assure the obj activity of the test program. These are common engineering practices and are appl icable to the Shuttle program. 21
Letter to the Honorable James C. Fletcher 8 The Panel expressed concern in its last report over the limited capacity of full scale test facilities. We believe that the addition of another full scale, full duration firing test stand capable of simulating dynamic loads is a major step forward in that regard. Conditions that should be simulated on the stand include launch loads and the most severe fit ight conditions. We have recently learned, however, that the new facility may be limited in its capability to simulate these loads and further that estimates of the maximum loads are being revised downward. We are concerned both about how well the loads are understood and the adequacy of the new facility. We recommend that the design of the stand be carefully reviewed to assure that the test results will be most useful. NASA apparently has concluded that another Joint Environ- ment S imulator (JES) test stand is not needed because addi- tional test hardware will allow more frequent tests on the existing stand. Each of the three JES tests conducted to date (using case joints of the original design) yielded results that were not predicted by prior analyses. For example, tests have shown that the extent of joint gap opening depends on whether pressurized gases reach the region between the clevis leg and the tang. The JES-IB test showed that the silicone O-rings cannot be used with the current grease. JES tests also revealed that pressurized gases can leak past the seals significantly before the occurrence of the peak pressure or peak gap opening. These tests have provided extremely valu- able data and insight into the behavior of joints. They also supply data to improve the analytical models. We concur that additional test hardware for use in the current JES and the related Transient Pressure Test Article will enhance the full scale test capability. If the rate of timely testing in these facilities is not sufficiently enhanced by addition of test hardware, we recommend that NASA reconsider its decision not to build an additional stand. At this writing, NASA has not formally approved the firings of Qualification Motors 7 and 8 before flight. In view of the number of features of the new design that depend critically on full scale, full duration testing, the Panel recommends that these be approved and incorporated into the test program immediately. Furthermore, the Panel believes that the dynamic tests in full scale, full duration firings with structural deformations to simul ate launch and flight loads must be completed before f ~ ight . 22
Letter to the Honorable James C. Fletcher Finally, we are grateful for the received from NASA, Thiokol, and redesign team deserves praise for work. On behaI f of the Panel, I remain , cc: Adm. Richard H. Truly Panel Members _9 _ cooperation that we have other personnel. The its dedication and hard S incerely, H. Guyford Stever Chairman
