Abstract: Wire breakage is the most serious quality accident which leads to the expensive materials scrapped in multi-wire sawing. How to select reasonable process parameters assuring wire reliability is an urgent problem. In this paper a method for assessing wire net reliability is presented. Assuming the fracture strength distribution of the worn wire is accordance with Weibull distribution, Weibull parameters of the samples is obtained using linear regression method and function relation of the wire reliability between wire tension and wire wear loss is derived. Then the wire web series discrete model is built. Furthermore wire web reliability is evaluated combining reliability theory. In order to verify the actual effect of this method, the experiments are conducted to verify the method. The experimental results show that the Weibull distribution can better describe the fracture strength distribution of the worn wire samples. There is a complicated nonlinear relationship between the wire net reliability, the wire tension and the wire net maximum wear loss. When wire net maximum wear loss is 4 μm, 7 μm, 10 μm, if the tension is less than 25 N, the difference of wire net reliability is small. Otherwise the wire net reliability decrease as parabola type with the increasing tension. Wire net reliability drops faster as more wire net wear loss. When wire tension is 20 N, 25 N, 30 N and wire wear maximum loss from 2 μm to 10 μm, wire net reliability drops with different types. The curves like horizontal line, steady slope line and plunge dropping line respectively. This method provide the theoretical rule for determining the reasonable slicing process reducing the risk of wire breakage.

Key words: multi-wire sawing, Weibull distribution, wire breakage, wire net reliability