The efficiency of overlapping joints in copper and aluminium busbar conductors

S.W. Melsom, H.C. Booth
1922 Journal of the Institution of Electrical Engineers  
The object of the investigation was to determine the amount of the contact resistance between the surfaces of an overlapping joint in rectangular straight busbar conductors of copper and aluminium under various conditions of surface preparation and pressure, and hence to ascertain the efficiency of such joints. An efficient joint was regarded as one in which the total resistance, which includes effects due to stream-line distortion, does not exceed that of an equal length of continuous
more » ... continuous conductor on either side of the joint. As a preliminary it was necessary to determine and allow for the effect of stream-line distortion under various conditions of overlap. Generally full-sized models were investigated, and the two surfaces with varying amounts of overlap were pressed together by means of an hydraulic ram or by bolts or special clamps. Numerical results are deduced and tables and graphs given for the contact resistance per unit of surface area under various conditions of pressure and amount of overlap, and after various modes of clean'ng and preparing the surface, including the interposition of tinfoil or the soldering of the two bars together. A series of graphs are given which show the efficiency of copper and aluminium joints in terms of the above defined standard for a wide range of values of busbar sizes and for any value of specific contact resistance likely to occur under practical conditions, and hence the amount of overlap required to obtain an efficient joint. The efficiency of making joints by means of clamps and bolts was also considered and the pressure occurring at such joints estimated by comparison of the contact resistance obtained when the same joint was held together in the hydraulic press. The effect of modifying the numbers, size and position of the bolts used was also investigated. The typical form of a clamped joint between flat conductors of rectangular cross-section is shown in Fig. 1 . The data required should enable the resistance of the overlapping portion A x A 2 to be calculated, and the ratio of its resistance to that of an equal length of single conductor of the same cross-section and resistivity to be predetermined. The problem under consideration divides itself into two heads : (1) The increase of resistance due to the streamline distortion which occurs at the overlap. (2) The contact or transition resistance between the superposed metal surfaces, and the manner in • The Papers Committee invite written communications (with a view to publication in the Journal if approved by the Committee) on papers published in the Journal without being read at a meeting. Communications should reach the Secretary of the Institution not later than one month after publication of the paper to which they relate. which this is affected by pressure and area of contact, and the condition or mode of preparation of the surfaces. (1) THE EFFECT OF STREAM-LINE DISTORTION. If the length of overlap is large compared with the thickness of the bar, the metallic resistance of the ovei lapping portion will obviously be very nearly equal to that of two conductors in parallel. But if the overlap is much reduced this ceases to be strictly true, as, owing to the distortion of the stream lines, the resistance is greater than that of two bars in parallel. It will be obvious that this distortion depends only on the ratio of the overlap to the thickness of the bar, and, provided the two bars are of equal width, it will .fl>i Ai 1 i. As a *F IG. 1. not be affected by the width ; the current-flow can therefore be regarded as taking place in two dimensions only. To determine the magnitude of this effect a series of experimental observations were made by means of a tinfoil model. A piece of thin, uniform tinfoil was cut into a shape which represented the longitudinal section of an overlap joint (Fig. 1). A steady current of suitable strength was passed through the strip, and the voltage-drop v± between the points Ai, A 2 was FIG. 2. compared by means of a potentiometer or millivoltmeter with the voltage-drop r 2 between the two points A 3 , A4, placed an equal distance apart in the straight portion of the strip. Since the voltage-drop is proportional to the resistance, the factor a = Vi/v 2 expresses the increase in metallic resistance due to stream-line distortion. This value is clearly constant whatever the dimensions of the overlap, provided that the ratio of the length I to the thickness t remains the same. To determine a for other values of the overlap ratio l/t, the tinfoil was cut away as indicated by the dotted lines in Fig. 2 . In this way a series of values were obtained for the ratio l/t, and for each of these values the resistance ratio a was determined. A sufficient number of
doi:10.1049/jiee-1.1922.0068 fatcat:dlta3kvpv5amdc2b6foooaqmse