Publications

This paper develops Classical and Bayesian methods for quantifying the uncertainty in reliability for a system of mixed series and parallel components for which both go/no-go and variables data are available. Classical methods focus on uncertainty due to sampling error. Bayesian methods can explore both sampling error and other knowledge-based uncertainties. To date, the reliability community has focused on qualitative statements about uncertainty because there was no consensus on how to quantify them. This paper provides a proof of concept that workable, meaningful quantification methods can be constructed. In addition, the application of the methods demonstrated that the results from the two fundamentally different approaches can be quite comparable. In both approaches, results are sensitive to the details of how one handles components for which no failures have been seen in relatively few tests.

Abstract not provided.

To determine whether a component is meeting its reliability requirement during production, acceptance sampling is employed in which selected units coming off the production line are subjected to additional environmental and/or destructive tests that are within the normal environment space to which the component is expected to be exposed throughout its life in the Stockpile. This report describes what these tests are and how they are scored for reliability purposes. The roles of screens, Engineering Use Only tests, and next assembly product acceptance testing are also discussed, along with both the advantages and disadvantages of environmental and destructive testing.

In performing assessments of low probability, high consequence systems, it is often preferable to use more than one methodology in order to assure that such systems undergo a thorough assessment. Hence, employing two methodologies in a complementary manner allows the analyst to bring the strongest features of each approach to bear upon the problem. The results of one methodology can be used to crosscheck or better characterize the results of another methodology, with the results being synergized in providing a comprehensive assessment of the system. This paper will briefly describe both the first principles and model based safety assessment methodologies, and will illustrate how both methods are used in a complementary manner in order to perform overall safety assessments of low probability, high consequence engineered systems at Sandia National Laboratories.

One of the major outgrowths resulting from the collapse of the former Soviet Union (FSU) has been an increase in technical information exchange and dialogue between the Russian and American nuclear weapons laboratories. One area of such discussions is concerned with the safety of low probability, high consequence systems and operations. In order to further the understanding between the respective institutes in this important area, a collaborative effort has been established between Sandia National Laboratories and the two premier Russian nuclear weapons laboratories, Arzamas-16 and Chelyabinsk-70, in which a common database has been developed which contains safety information provided by all three laboratories. More than 1,200 documents have been placed by the three institutes into this database. This paper describes the details of this data base, including the types of safety information being stored.

One of the benefits resulting from the collapse of the Soviet Union is the increased dialogue currently taking place between American and Russian nuclear weapons scientists in various technical arenas. One of these arenas currently being investigated involves collaborative studies which illustrate how risk assessment is perceived and utilized in the Former Soviet Union (FSU). The collaborative studies indicate that, while similarities exist with respect to some methodologies, the assumptions and approaches in performing risk assessments were, and still are, somewhat different in the FSU as opposed to that in the US. The purpose of this paper is to highlight the present knowledge of risk assessment methodologies and philosophies within the two largest nuclear weapons laboratories of the Former Soviet Union, Arzamas-16 and Chelyabinsk-70. Furthermore, This paper will address the relative progress of new risk assessment methodologies, such as Fuzzy Logic, within the framework of current risk assessment methods at these two institutes.

An ever increasing demand for highly rugged, miniature AT strip resonators prompted the development of a resonator package for use in high-g shock applications. This package, designed and developed by Statek Corporation, is based on the package configuration currently being used by Statek for commercial devices. This report describes the design intent, component characteristics, and evaluation test results for this device.

Salford Electrical Instruments, Ltd., and the General Electric Company`s Hirst Research Centre, under contract to the United Kingdom`s (UK) Ministry of Defence, developed a radiation-hard, leadless chip-carrier-packaged oscillator/divider. Two preproduction clocks brought to Sandia National Laboratories (SNL) by a potential SNL customer underwent mechanical and thermal environmental evaluation. Because of the subsequent failure of one device and the deteriorating condition of another device, the devices were not subjected to radiation tests. This report describes the specifics of the environmental evaluation performed on these two clocks and the postmortem analysis of one unit, which ultimately failed. Clock startup time versus temperature studies were also performed and compared to an SNL-designed clock having the same fundamental frequency.