Fabrication, Distortion, and Metrology of Shrink Fit Electrical Connections

Abstract

The fabrication of pulsed power experiments on systems such as ATLAS at Los Alamos National Labs requires assembly of shrink fit or press fit joints. These joints are used to carry large electrical currents on the order of several megamperes for Z-pinch cylindrical implosions of thin-walled cylinders. The joints, which represent the shrink fit or press fit interfaces between cylindrical components (1100 aluminum liners and copper glide planes), are also called on to support mechanical stresses. As a result, the integrity of these joints is critical to the success of the mission. The analytical challenge for investigating these components lies in the presence of a partial shrink fit, that is, a shrink fit in which the length of contact between mating cylinders does not equal the entire length of at least one of the components. Such a design leads to stress distributions that are impossible to calculate using standard shrink fit theory that is analytically tractable. Arguably more important, however, is the varying radial deflection profile along the length of the liner that results from the partial shrink fit. For the success of these experiments, uniform implosions are deemed to be highly beneficial; thus, determining how to predict and compensate for liner deflection profiles is crucial.

The primary issues at hand include the following: mechanics of the interference fits, physical description of the contact surfaces between the liner and glide planes, joint void (gap) description, and material property effects. Furthermore, for the original shrink fit design of ATLAS, liner geometry alterations were investigated for potential performance improvement, particularly for the elimination of joint interface gaps and lower stress magnitudes. For the modified ATLAS design that included components designed to compress and distort other elements, effects of localized geometry modifications were investigated for the same purposes. Composite liners, i.e., liners with two constituent materials, were also examined, along with potential problems such as cylindrical out-of-roundness on the precision scale. To investigate these issues, finite element analysis models have been created and compared to experimental data. A measurement technique has been developed to compare radial deflection profiles of diamond turned thin-walled shrink fit test specimens to finite element models of corresponding geometry. Verification of the computer generated finite element models with experimental results could provide tremendous aid in accurately predicting the shape, deflection, and stress distribution of such cylindrical elements.