Investigations on precision finishing of helical gears by electrochemical honing process

Abstract

This paper presents the findings of an investigation into the effects of finishing time, electrolyte-related parameters (i.e. electrolyte temperature, composition, and concentration), voltage, and rotating speed of the workpiece on the improvement of surface quality of helical gear teeth finished by the electrochemical honing (ECH) process. Percentage improvement in average surface roughness PIRa and maximum surface roughness PIRtm, three-dimensional plots of the finished surface texture, and microhardness have been used as the measures of ECH process performance. The effects of finishing time, electrolyte temperature, and electrolyte composition have been studied through pilot experiments by varying one variable at a time, while the effects of voltage, rotating speed of the workpiece gear, and electrolyte concentration have been studied during the main experiments designed using the Box–Behnken response surface methodology approach. Scanning electron microscopy and optical profilometry before and after ECH have been used to highlight the improvement in surface quality of helical gears. A special type of cathode gear has been developed by sandwiching a conducting copper gear between two insulating gears of Bakelite, having the capability of varying the rate of electrolytic dissolution steplessly along the full profile of the workpiece gear made of EN8. Based on the results of pilot experiments, 7.5 min as finishing time, a mixture of NaCl and NaNO3 in a ratio of 3:1, and 32 °C as electrolyte temperature were found to be optimum for precision finishing of helical gears. Using the results of main experiments, regression models have been developed for the measures of process performance (i.e. PIRa and PIRtm). Analysis of variance done at the 95 per cent confidence level, to test the significance of the developed models and process variables, found that the developed models are highly significant and that voltage and electrolyte concentration have significant effects on the responses. However, no significant interaction effect has been observed. Predictions from the regression models have been validated by comparing them with the results of the confirmation experiments, which proved that the developed models are correct and acceptable.