No abstract

No abstract

The stainless steel-304 is used in various surgical instruments due to its low carbon content and better corrosion resistance. SS-304 is machined using photochemical machining process (PCM) to obtain the components with micro-dimensions. The statistical methods of signal-to-noise (S/N) ratio and the analysis of variance (ANOVA) are applied to investigate effect of concentration of etchant, time of etching and temperature of etchant using DoE (L 27) orthogonal array. The performance characteristics like material removal rate (MRR), surface roughness (R a), undercut (U C) and etch factor (EF) are optimized during the investigation. The ANOVA technique is used to evaluate significance and percentage contribution of each parameter. The regression model for PCM of SS-304 is developed. The ANN technique is used to compare the predicted and experimental results of machining. The MRR, R a , EF and U C are showing improvement of 0.38 mm 3 /min, 1.271 µm, 0.11 and 0.028 mm, respectively, after the confirmatory test. The developed model is used to manufacture the array of micro-holes (5 × 5) with hole size of 220 µm, 280 µm and 370 µm which are used for metering of medicine in biomedical applications.

Non-conventional process like Photochemical Machining (PCM) is found to show a promise for machining very thin metal components. In the present study, the effect of various selected parameters such as time of etching, temperature of etchant and concentration of etchant on material removal rate, undercut in PCM of phosphor bronze has been investigated by using multi-objective grey relational analysis and their optimal conditions are evaluated. Full factorial (L27) orthogonal array (DoE) has been used to perform the experiments. GRG value indicates most significant parameters affecting the PCM process. The above factors are selected on the basis of effect - cause analysis and literature survey. Mathematical models relating to the machining performance and machining parameters have been formulated. Optimal settings for each performance measure have also been obtained. The results obtained after conference test prove that improvement in the quality will take place is if the setting of parameters are done at optimum level predicted by multi-objective grey relational analysis. The ANN model is prepared to predict the result by training neural which can be compared with actual experiments to confirm the satisfactory performance during the experimentation.

No abstract