The use of strain gauges in the prediction of sidewall cutting forces, in milling
ABSTRACT:
This study investigates the use of strain gauges in the prediction of cutting forces in slab milling of aerospace specification aluminium. Can strain gauges be used in the prediction of machining forces? An experiment is furmulated using Minitab (in order to deal with a large number of variables). Captured strains are validated using a secondary test, involving the Instron universal tensile tester. Forces are calculated from the recorded strains using bending moment equations and the process is simulated using the Instron machine to try to validate these calculated forces. Numerous techniques are used and assumptions made in order to specify the problem. This thesis analyses the problems involved, advices on their solutions and most importantly highlights through graphical representation how well strain gauges can perform in these conditions.
INTRODUCTION:
This theisis is a study of the sidewall forces in milling. The purpose of the study is to try and find a method for predicting these forces. Prediction of these forces is of major importance as it helps significantly in reducing manufacturing times and costs (rework and scrap). Engineering/Manufacturing companies often have trouble when it comes to machining thing cross-sections. Airbus and the aerospace industry as a whole are particularly interested in methods of force prediction as they regularly experience difficulties in achieveing optimum dimensions for the designs of their airplane wing ribs. Once the cross-section of the rib reaches a certain "thickness" they deform under the forces applied by the cutter. It is at this thickness that the designers must plan around. If there were some technique that would allow engineers to know what forces are applied during cutting, they could have fewer limitations imposed on their designs.
The ideal solution would be to input all the relevant variables into an equation or mathematical model, for example, dimensions of the specimen, cutter material, depth of cut, cutter speed etc, which produces a result for the output force experienced by the sidewall of the component as a result of those variables. Such an equation would allow optimum dimensions to be achieved at the design stage, reducing rework and scrap materials resulting in an overall decreased cost and a reduction in manufacturing times. The prediction of these forces cannot be estabilished from one experiment or even in the amount of time available to complete this thesis. The aim here is to test some of the variables to a suitable degree of understanding in order to evaluate the process and hopefully find a link between the variables and the forces exerted on the component. This study will take future researchers in this field a step further to understanding the connection between all the variables involved in the milling process, by attempting to establish, which of the three chosen variables (speed of cutter, aluminium cross-section and cutting distance from gauge) has the largest effect on the output forces. The thesis will also try and prove that the technique used to measure the strain is a valid one. The machining operation under review is the "milling operation". This cutting technique was chosen as it is the most widely used in industry due to its versatility and also because of its use in the production of wing ribs at Airbus. It is for the same reason that the material chosen to undergo machining is aerospace specification aluminium.
The milling operation is a straightforward process but the fact that it can be used in various orientations and speeds makes it difficult to analyse (these orientations and other variables are discussed in sections 3.1 and 4.1.1 of the thesis) because any one slight change made to the process can potentially have major affects on the forces that are exerted on the specimen. It is obvious therefore, that in order to obtain a full understanding of the output forces, every parameter must be tested individually and an experiment must be conducted for every possible variable to see which of them have the most substantial effect on the resultant force. To do this would require a huge amount of time, which simply wasn’t available to me. Therefore, to get around the problem of having many variables some were to remain fixed depending on valid reasoning. These will be discussed further in section 4 of this Thesis.
