FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1956
1956 - 1083.PDF
FLIGHT, 10 August 1956 229 COMPONENT RELIABILITY Relative Complexity Analyzed By L. F. E. COOMBS THE maintenance and supply requirements of a combat air-craft are largely dependent upon the reliability of themachine itself, which in turn is dependent upon the relia- bility of its components. For the purpose of an analysis, it canbe assumed that the aircraft structure has little effect upon the problem. The components which influence reliability most willbe the "black boxes," such as electric and electronic radar, fire control and power control units. It should be noted, however,that the reliability of electronic units may be improved through the introduction of printed circuits and transistors as substitutesfor the conventional "tube" circuits so that the key units to overall reliability might then be the units which contain rate-gyro instruments. Reliability must always be one of the foundations of the designof an aircraft component or system of components—the more so at the present time, when each component depends upon thecorrect function of others. Unfortunately, under the stress of war (or cold war) conditions reliability becomes sacrificed toexpendability and_ availability. In any case, it is not so easy to draw a dividing line between a standard which gives a long lifewith reduced maintenance and low availability and one which gives a short life, but requires a large maintenance rate. Availability, as far as combat aircraft are concerned, is afunction of production rate, the logistic efficiency and the facility with which a failed unit can be replaced by a new unit. Reliabilityof components has to be a compromise between the requirements of the operator and the production requirements of the producer,who—especially if called upon to supply large quantities—will want to produce a unit with a comparatively short life, guaran-teed by a large-scale inspection system to within a narrow mar- gin. The operator will call, naturally, for a unit which has thelongest possible life and therefore the minimum maintenance requirements. Each side will then have to adjust its ideas inrelation to the dictates of its financial controllers. As an example, a unit could be made to design Type A, whichwould be guaranteed by the producers to have a mean life, with- out failure, of 1,000 hours at a cost of £100. It would be capablealso of being "plugged" straight into the aircraft without the need for adjustments, or for calibration to suit individualdeviations from the mean airframe characteristics. As an alternative to Type A, the producer might offer TypeB, which would have a mean life of 50 hours and cost £10. However, apart from other factors, Type B would need to befitted into the aircraft by skilled personnel and require time for calibration and other adjustments. Besides having to take such facts into consideration the opera-tor has to consider the availability and the problems of transport and storage of replacement units. In the case of the Type Aunit the supply rate might not be high enough to match the expected replacement rate, even though the replacement ratewas less than that for the Type B. In contrast, the replacement rate for Type B, which for purposes of comparison is assumedto be a function of the mean life, is twenty times the rate for Type A. Such a rate will require approximately twenty times thetransport capacity and twenty times the storage capacity of Type A. Also to be considered is the failure rate during the develop-ment period—extending in most cases into the operating life of the aircraft. The Type A unit has, owing to the requirementsof a long life and ease of replacement, a greater number of parts than Type B. During the development period the reliability ofType A, a function of the individual part reliabilities, will be possibly lower than the less complex Type B. Under ideal conditions both kinds of units would be deliveredwith all the development snags eliminated. Unfortunately, that is not the usual procedure, and as a result the operator mustconsider not only the length of the development period, but its length in relation to the normal expected life after all snags havebeen eliminated. How much development time can be allowed? It is possible that the more complex unit will suffer from morefailures during the development period than the simpler "short- life" type. The curves illustrate the different types of "life." Fig. 1 illustrates the relationship between reliability and life.Curve A, for the "long-life" type of unit, shows that initially the reliability is lower than that achieved by using Type B (curve B).If the minimum acceptable reliability level is known, the inter- ACCEPTABLE LEVEL OF RELIABILITY COMPARATIVE NORMAL LIVES Fig. 7. -FUNCTION OF ^COMPLEXITY TIME- Fig. 2. RELIABILITY OF COMPONENTS 10 20 30 40 50 60 70 80 90 100 NUMBER OF PARTS r. ': Fig. 3 (above). Weakest link—an illustration oi the manner in which the overall reliability of a unit is influenced by the degree of reliability of its component part. The number of parts shown is arbitrary. TIME »• Fig. 4. Figs. 1, 2 and 4 (left). Simplicity or complexity? The diagrams com- pare the development and life advantages of units composed of greater or fewer numbers of parts. The inference of the function of complexity is discussed in the text.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
