Experimental investigations of significant parameters of strain measurement employing Taguchi method

Abstract

Designing a safe load-carrying component for mechanical structures requires sufficient information related to the distribution of loads, which is applied to the structure. One of the main design criteria is the level of stress. In some test cases, measuring the stress directly becomes very difficult due to a complex geometry for instance. In this case, the measurement of the physical displacement of the structure, which is known as strain (ε) facilitates this procedure. The accurate strain gauges are mostly sensible and several factors can affect the measurement performance. It is obvious that the resistance tolerances and strain that induced during the application of the gauge, generate some initial offset voltage without the presence of forces. In addition, long lead wires can add resistance to the arm of the circuit bridge, which adds an offset error and desensitizes the output of the bridge. Hence, understanding the effects of significant parameters for analyzing the data is considered valuable. In this paper, the influence of the temperature, the length of wires and the set point of forces on the strain measurement are experimentally investigated at three levels. Furthermore, the design of experiment and related tools such as Taguchi and signal to noise ratio are employed to reduce the errors and uncertainty through the measurement process. Noise can be related to controllable factors or uncontrollable factors, which are extraneous factors. Using the Taguchi method assists to provide a better control over the right set of controllable factors only, to determine the significant factors with their optimal levels. The results indicate that the temperature has a maximum contribution to the measurement of the strain with 76.35%, which is followed by the effects of length of wires and the point of application with 7.66% and 6.77%, respectively.