FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1961
1961 - 0843.PDF
FLIGHT, 22 June 1961 855 ij-velop and improve component reliability was put into effect.While overall expenditures might be higher in the early part of the programme, he concluded, the total amount of money spent by theend of the programme would be less (see reliability and cost curves below). As for what made reliability easier to gain in one propulsionsvstem compared with another, Dr Ritchey listed two major items. First, the less reliable system might have more components, thusincreasing the number of separate elements of the system that could gn wrong. A more complex system would require a more extensivetesting programme in the development phase. Second, the less reliable system might have been designed closer to the limits ofexisting knowledge so as to achieve the required performance, and therefore would include a higher proportion of unknown factors. In a frank claim for the greater reliability of rocket engines em-ploying solid propellants, Dr Ritchey cited the reliability history of three different rocket powerplants—the XM-20 solid-propellantengine used as the first stage of the XI7 re-entry test vehicle, the XM-55 solid-propellant engine to be used in the first stage of theMinuteman missile, and the LR-89 liquid-fuel engine used in pairs in Atlas (see Table I). Dr Ritchey showed how the XM-20, with only six components,achieved a demonstrated reliability of 95.7 per cent after only six pre-flight rating tests. "Following these six static tests," said DrRitchey, "a total of 39 flights was accomplished with no recorded DOLLARS SPENT. PROGRAM Ml TABLE I: RELIABILITY RECORDS FOR THREE ENGINES XM-20 XM-55 (XI7- ! MM- RTV) | Wing I LR-OT Atlas (1) Number of components (2) Number of complete engine development tests ' (a) Engineering development and storage ' tests (b) Preflight rating tests ... (c) Qualification tests (d) Development flight tests (3) Accumulated running time to date (4) Total flight tests to date (5) Flight failures due to 1st stage propulsion ... (6) Reliability demonstrated (90% statistical confidence) I 5 0 0 1,220 sec 39 0 95.7 9 132* 82* 12 18* 20* 2,350 sec 1 0 33 3,177 >3,I26 J 51 155,750 sec 74 4 90.5 91.4 DOLLARS SPENT. PROGRAM <>! I I RELIABILITY GROWTH "A" I I RELIABILITY GROWTH "B" [ I ^- FIRST SYSTEM TEST >" END °F PFRT | * Projected to completion of first operational wing. The above compilation of test data for two solid-fuel rocket engines (XM-20 used in the XII re-entry test vehicle and the XM-55 for first- stage Minuteman) and one liquid rocket engine (LR-89 in Atlas) show how many more tests must be run on more complex engines to achieve reliability compared with a simpler engine. The XM-20 with six com- ponents required only six complete engine development tests to achieve a reliability of 95.7 per cent, whereas the XM-55, with 50 per cent more components, will require 132 tests to achieve an estimated 90.5 per cent reliability. The LR-89, on the other hand, with 33 components, has taken 3,177 complete engine development tests to turn in a performance of 91.4 per cent reliability RELIABILITY AND COST CURVES Two different but typical situations are depicted by curves A and 8. Both show an improvement in reliability over a period of time Reliability growth curve A might be based on a simple system with a minimum number of components, or a system designed rather conservatively to assure performance. B, in which lower reliability is experienced, may have more components than A, or possibly B had to be designed more closer to the limits of existing knowledge in order to obtain necessary performance. The upper curves depict the manner in which money may be spent during the course of a particular development pro- gramme. Curve I is linear, and shows spending during the course of a development programme relative to the degree of reliability attained in either System A or System B. Curve 2 shows spending of greater amounts early in a programme, primarily for extensive testing to develop and improve com- ponent reliability. Because of these early outlays, the same degree of reliability is attained by the end of the programme with a smaller outlay. Together, these charts show that almost any degree of reliability can be attained at a cost in time and money, and that such costs must be weighed carefully against the need for a higher degree of reliability in the system. END OF QUALIFICATION failures attributed to the propulsion system. Since that time, thisengine with certain modifications has been used extensively in the Polaris test vehicle programme, the NASA Little Joe and Scoutprogrammes, and in the Argus launches. More than 100 flights have been launched with no propulsion failures." The XM-55, with nine components (50 per cent more than theXM-20) would go through 132 complete engine development test flights and achieve a reliability of 90.5 per cent by the time oneoperational wing of Minuteman missiles had been completed, Dr Ritchey estimated. The LR-89 liquid-fuel engine for Atlas had achieved a reliabilityof 91.4 per cent to date, but had required 3,177 complete engine development tests to demonstrate that degree of reliability. TheLR-89, as noted previously, had 33 components, "and as compo- nents increase," declared Dr Ritchey, "the number of tests mustgo up geometrically to achieve the same degree of reliability as for an engine with fewer components." Dr Ritchey noted that, in the XM-55 engine development pro-gramme for Minuteman, "There have been a number of compo- nent failures consistent with the usual experience in solid enginedevelopment" and with the fact that the engine had a high degree of design sophistication. Once satisfactory designs were achieved, nosignificant failure was experienced in 23 subsequent tests. To date, some 178 tests had been run on the combustion chamber of theXM-55 engine for Minuteman, 196 on the nozzle, 1,260 on the liner and insulation, 2,125 on the ignition system, and 57,756 onthe propellant. The large number of propellant tests included numerous small engine tests for propellant ballistic and environ-mental evaluation, physical property tests on propellant samples, (Concluded on page 858) The other side to the coin: failures are inevitable in development, and sometimes spectacular (fifth Titan, August 14, 1959)
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
