11- 3.1 Introduction

(a) Importance of Design Phase. Reliable instrumenta­tion systems can only be created during the design phase. No amount of quality control, field testing, or maintenance can adequately compensate for a lack of careful planning during the conceptual design. A few designers seem to have an intuitive sense of good design, and their products enjoy a reputation for reliability Most designers can acquire the art of producing reliable designs by adhering to some of the disciplines of reliability engineering

A designer’s most frequent failing is to be trapped into thinking only about how to design a system that will work and giving no thought as to how it might fail. Everyone is amused by the famous Rube Goldberg designs (Fig. 11 1) It is readily apparent how Rube’s systems work and equally apparent how prone they are to fail. Modern reactor instrumentation systems have become so sophisticated that the designer, in concentrating his efforts on making the system work, forgets to look for ways in which it may fail By applying the techniques of reliability analysis, a good designer should consistently turn out a product that not only works but also has a low incidence of failure

The attitude of the designer is all important He must have the desire to create a reliable system. He must not be lulled into the attitude that the design is adequate because the system passed a design audit or was granted an operating license. A trained reliability engineer can probe deeply and learn much about a system, but, unless he has full cooperation from the designer, the effort will fall short of complete success

System reliability and system capability must not be confused. For example, the nameplate rating on a power supply tells the designer about the load capacity it may be expected to accommodate. Choice of a power supply with a rating in excess of the load requirements assures capability. A reliability study assumes system capability and goes on to make an assessment of the probability that the system will actually be successful in performing a given task within its capability The two terms, capability and reliability, are related through the term “design margin,” and the designer should recognize the favorable influence excess capability may have on reliability through the application of appropri­ate derating factors.

Effort invested in systems reliability analysis is eco­nomically attractive The potential for cost saving lies in systematically selecting the higher reliability systems for detailed design, in choosing the simpler of two alternatives, in avoiding overdesign on portions of the system that do not contribute to reliability, in avoiding costly retrofits, and in gaining a reputation for a reliable product

(b) Reliability and Availability. The term “reliability” is frequently used in a qualitative sense to imply quality

and integrity As pointed out m Sec 11 1 of this chapter, reliability is a measure of the time stability of a product’s performance This concept can be expressed in a quantita tive sense in the definition of reliability approved by the IEFL 4 “ 1 he characteristic of an item expressed by the probability that it will perform a required function under stated conditions for a stated period of time ” In the general sense, this definition of reliability does not allow for failure and repair during the stated period of time For example, consider a pressure switch monitoring the reactor pressure located where it is totally inaccessible during reactor operation An assessment of the numerical reliabil­ity of the switch to survive one fuel cycle is a meaningful measure of its integrity

In many other cases, inspection, test, and repair during operation are permitted, and this is, in fact, the preferred mode of operation In this event, the most meaningful measure of integrity is called “availability” and is defined as4 “The characteristic of an item expressed by the probability that it will be operational at a randomly selected future instant in time ” Stated another way, availability is the fraction of the time the system is operational In the long-term steady-state situation, avail­ability is given by the equation

Up time

Up time + Down time where the up time is approximately equal to the average time between failures and the down time is the average time to repair and restore the system to service Since the down time is the average time to repair the system, it must include the elapsed time between the failure and its discovery, a term that may be of primary significance especially in standby systems

Availability, not reliability, is the term that is most applicable to the usual reactor instrumentation system, where redundancy, repair during operation, and a para­mount concern for detecting and eliminating unsafe failures are prominent characteristics