242 PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 89 NR 2a/2013 Peter DINEFF, Valentin DIKOV, Raina TZENEVA Technical University of Sofia Stress Relaxation in Copper Bolted Busbar Assemblies with New Design Concept of the Bolt Holes Abstract. This article reports experimental investigation results of the stress relaxation in copper bolted busbar assemblies with new design concept of the bolt holes (S-design and G-design). The results of the new case are compared with the results of the classical case of the same assemblies and it is observed that for the new S- and G- designs the decrease in the fastening force is quite low which stands for a higher level of reliable assembly operation. Streszczenie. W artykule opisano badania eksperymentalne, dotyczące naprężeń w zespole miedzianych szyn zasilających o połączeniach śrubowych. Zastosowano nową propozycję otworów śrubowych, które, po porównaniu z klasycznymi metodami, wykazały mniejszy spadek siły docisku, co przekłada się na zwiększoną niezawodność połączenia. (Relaksacja naprężeń w zespole miedzianych szyn zasilających o połączeniach śrubowych – nowy rodzaj otworów śrubowych). Keywords: bolted busbar assemblies, experiment, stress relaxation, bolt hole design, loading sequence, regression equation. Słowa kluczowe: szyny zasilające o połączeniu śrubowym, eksperyment, relaksacja naprężeń, projekt otworów śrubowych, sekwencja obciążeń, równanie regresji. Introduction One of the most important concerns in the design of bolted busbar assemblies is the initial contact force. Considerable effort is required to provide the right holes geometry and the right penetration, and to select an alloy with the necessary strength and adequate formability. But these initial efforts must not be the only concern because some metals lose a significant amount of their initial contact force over time. This force decay is a natural consequence of the structure of all metals and is called stress relaxation. Stress relaxation is the time dependent decrease of the stress in a metal under constant strain, such as in a busbar with fixed deflection. Pure copper has poor stress relaxation resistance. Adding different elements to copper produces alloys with different combinations of strength, conductivity, formability- and stress relaxation resistance. All of the copper alloys are susceptible to stress relaxation, just as are all other metals. But there is a tremendous difference in the stress relaxation between different copper alloys, so alloy selection can be important. In the strictest sense, stress relaxation always occurs, over time, in any stressed member of a bolted busbar connection. But in a practical sense, some applications will experience very small losses of contact force. For highly resistant alloys, under moderate stress for short times, without exposure to high temperatures, the decrease in contact force will be within the experimental error of pre-production testing. By understanding the variables that affect stress relaxation the designer can make choices which optimize the behavior of a contact during its expected life. Once the metal (or alloy) choice has been made, there are two other important variables that affect stress relaxation: time and temperature. And there are three variables that have less of an impact: initial stress, orientation, and the degree of cold work. The latter is expressed as temper, a traditional designation which reflects the cold reduction by rolling in cold rolled alloys, or the mechanical properties in the case of certain heat-treated alloys. Although time and temperature are usually determined by the expected application of a bolted busbar assembly, with a suitable choice of contact metal (alloy) the loss of contact force over design life can be curtailed. By management of three secondary variables, the loss of contact force over time can be further reduced. The secondary variables are: initial stress level, orientation, and temper. The rate of stress relaxation is influenced by the initial stress. If the busbar is initially deformed such that it is stressed at its yield strength, then rapid loss of contact force may occur. If the stress is only 20% of the yield strength, practically no loss occurs. Initial contact forces are such that applied stresses are commonly 30-90% of the metal's yield strength, and within that range the loss of contact force is almost invariant with S i . That is why stress relaxation testing is usually done with initial stress levels of 75% of the metal yield strength. Although stress relaxation continues over long periods of time, the rate is highest initially and decreases with increasing time. This makes it convenient to plot the data with time on a logarithmic scale against the percent remaining stress, as shown in the figure below for cartridge brass. At a test temperature of 75°C, a little over half the initial stress remains so almost half the contact force has been lost. Stress relaxation is a function not only of metal or alloy, but also of temperature, orientation, initial stress, and, of course, time. To prevent circuit integrity from being compromised over the design life of bolted busbar assembly the loss of contact force due to stress relaxation must be taken into account. When should the designer be concerned? Not only during metal or alloy selection because, as previously noted, there are significant differences among the design of holes used for connectors, [1, 2]. There have been proposed 3 new bolt hole design cases of high power bolted busbar connections in [1] – S- design with slotted bolt holes, SH-design with slotted bolt holes, ending with small holes, and G-design where groups of small holes are situated around the bolt holes, as shown in Figure 1. Fig. 1 Classical (CD) and new designs for bolted busbar assemblies: a) S-design; b) SH-design; c) G-design The finite element simulation tool ANSYS Workbench is used to study the mechanical changes, associated with the contact pressure and depth of penetration in the contact area between two busbars in a bolted busbar connection. A higher contact penetration increases both the numbers and dimensions of α-spots. This in turn expands the true contact area and decreases the contact resistance, which is a precondition for introducing a new hole-shape design for this connection. Moreover, there is a significant