EXPERIMENTAL INVESTIGATIONS OF ELECTROCHEMICAL HONING OF BEVEL GEARS

Abstract

Bevel gears are used to transmit motion/power between two intersecting shafts by means of gradual engagement of teeth and hence, gear teeth finishing plays a crucial role in overall in-service performance and to improve the life span of it. Traditional gear finishing processes such as gear grinding, gear shaving, gear honing, gear lapping, etc. are presently popular in industries. In these processes, material is removed by shearing action owing to severe plastic deformation caused by relative motion between tool and workpiece and therefore, the tool material should be harder than workpiece material. Thus, these processes are to some extent limited by hardness of workpiece material. In addition, these are costly and time consuming. These shortcomings necessitate the exploration of cost-effective and advanced gear finishing processes in which the energy sources used for material removal is quite different from the traditional processes. In last few decades, electrochemical honing (ECH) has emerged as a most potential hybrid micro-finishing technique. Research community has started to utilize the fundamental concept of the process for precision finishing of internal cylinders and gears. The present study discusses the precision finishing of bevel gears by ECH process to explore the effect of influencing process parameters on surface topography and surface integrity of machined surface; tribological performance of gear teeth profile and process capability. For the present study, an experimental setup with modular tooling system has been indigenously developed. With the assistance of modular tooling system, the developed machine setup can provide the versatility of running ECH, ECM and honing process in a single setup beside incorporating gears of different sizes with minimum setup changeover. The experimental investigation has been conducted into three phases: pilot experiments, main experiments and confirmation experiments. Pilot experiments have been carried out to study the effect of processing time, inter-electrode gap, voltage, electrolyte temperature, electrolyte pressure and flowrate, electrolyte composition on measures of process performance i.e. percentage improvement in average surface roughness (PIRa), percentage improvement in root mean square roughness (PIRq), percentage improvement in maximum surface roughness (PIRz), percentage improvement in bearing ratio (PIBr) and material removal rate (MRR). The surface roughness and bearing ratio values of the samples before and after the process are measured with a Wyko NT 1100 optical profilometer interfaced with Vision®32 software to find out the v percentage improvement in surface roughness values and bearing ratio. Twenty separate measurements have been taken on gear teeth profile along the face width of the gear and average value is used for further analysis. Material removal rate is measured by calculating amount of metal removed per unit processing time. The experimentation to find out the optimal values of processing time, inter-electrode gap, voltage, electrolyte temperature, electrolyte pressure and flowrate have been conducted following one-factorat- a-time technique and the optimal values are identified by graphical analysis. The experimentation to find out optimal electrolyte composition is carried out according to the Mixture D-Optimal Design and optimal value is found out by desirability analysis. Based on the results of pilot experiments, 10 min processing time, 0.25 mm inter-electrode gap, 30 V voltage, 35 0C electrolyte temperature, 15 l/min electrolyte flowrate and (80% NaCl + 20% NaNO3) electrolyte composition are found optimum. The main experiments have been carried out to explore the influence of input process parameters: current, rotating speed of workpiece and electrolyte concentration on process performance characteristics. The trial runs are designed according to the Box Behnken Design of Response Surface Methodology in which current, rotating speed of the workpiece gear and electrolyte concentration are varied in the ranges of 10-30 A, 60-100 rpm, and 5%-10% respectively. Analysis of variance and parametric optimization are carried out with the help of Design- Expert (Version: 6.0.8) software for analyzing the results in order to identify the optimum process performance characteristics with a particular combination of input process parameters. Current and electrolyte concentration have been found to have significant effect on ECH process performance characteristics while rotating speed of workpiece is found to have insignificant effect. The surface finishing and material removal rate are found to improve with the simultaneous increase in current and electrolyte concentration and attain higher value at current of 30 A and electrolyte concentration of 10%. However, it is evident that process performance characteristics initially improve with increasing rotating speed upto 80 rpm and then starts declining indicating the presence of an optimum value of rotating speed for ECH of bevel gears in between 60 rpm and 100 rpm. The input process parameters used in the study are independent of each other and as a result of this the interactive effects of input process parameters on process performance characteristics are found insignificant. Regression models for PIRa, PIRq, PIRz, PIBr and MRR have been developed which can provide flexibility to the process users in deciding potential application of ECH of bevel gears. It is obvious that the developed models are suitable in predicting the results with reasonable accuracy (3.37% prediction error for vi PIRa, 5.19% prediction error for PIRq, 3.81% prediction error for PIRz, 4.62% prediction error for PIBr and 6.12% prediction error for MRR). Multi-objective optimization has been carried out on the basis of desirability analysis and 30 A current, 79 rpm rotating speed and 10% electrolyte concentration are found optimum for desirability value of 0.997. In this study experimental investigation has also been carried out using pulsating power supply to explore the influence of pulsating parameters i.e. processing time for pulse assisted ECH (PECH), pulse-on time, pulse-off time and duty cycle on measures of process performance. 30 min processing time for PECH, 1 ms pulse-on time, 1 ms pulseoff time and 50% duty cycle are found optimum. Based on the results, it is obvious that that 50% duty cycle provide approximately 5 times higher value of process performance characteristics in comparison to the duty cycle of 10%. It is observed that PECH provides better surface finish but with lower material removal rate. A comparative study between electrochemical honed gear and gear finished by pure mechanical abrasion has been carried out to demonstrate the capability of the process in improving surface quality of gear teeth profile. It is found that in case of ECH the electrolytic dissolution enhances the process capability by improving surface quality and MRR. It is evident that PIRa, PIRq, PIRz, PIBr and MRR values of ECH are approximately 2.5 to 4.5 times higher than respective values of mechanical abrasion. The surface characteristics of electrochemical honed surface have been analysed and it is found that ECH improves the surface quality by removing the small pits, micro-burrs, feed marks and irregularities. No significant changes in surface/sub-surface microstructure and micro-hardness have been noticed. A significant reduction in bearing ratio value (near about 60%) has been observed in processed surfaces. ANN modelling of process performance characteristics indicates good agreement between neural network predictions and experimental values. It is believed that ANN requires much more number of experiments to form a well-structured model while in this study, ANN modelling has been found to work well even with relatively less date as data is statistically well distributed in input domain. A comparative study of ANN modelling and RSM modelling has been carried out and it is observed that the ANN process is best vii suited than RSM for modelling the process performance characteristics of ECH of bevel gears even with less number of experimental runs. The optimized values of input process parameters are given in tabular form at the end of the thesis to help the potential users of the process. Keywords: Advanced machining process (AMP); hybrid machining process (HMP); electrochemical machining (ECM); honing; electrochemical honing (ECH); pulse assisted electrochemical honing (PECH); bevel gears; Mixture D-Optimal Design; Box Behnken Design (BBD); analysis of variance; desirability analysis; surface integrity aspects; ANN modelling. Organization of Thesis Thesis has been organized in following seven chapters. Chapter 1: It presents brief introduction of advanced machining processes and hybrid machining processes. Electrochemical honing process and its constituent process i.e. electrochemical machining and honing have also been discussed. Chapter 2: It contains the comparative study of various gear finishing processes and comprehensive review of past research work. On the basis of literature review, gaps in past research work are identified and the objectives of the present research work have been outlined. Chapter 3: It describes the details of designed and developed experimental setup and the material selection criteria for different components of the machine setup. Chapter 4: It provides the detailed discussions about electrochemical honing process parameters, workpiece selection, electrolyte selection, parametric study to find out ranges and levels of input process variables and fixed parameters, process performance characteristics and design of experiments technique to plan the experimental runs. Chapter 5: It contains the experimental outcomes of pilot, main and confirmation experimentation and analysis of the outcomes to investigate the optimum parametric combinations for both electrochemical honing and pulse-electrochemical honing of bevel gears to obtain the better improvement in surface quality and higher value of material removal rate. viii Chapter 6: It describes the artificial neural network modelling of electrochemical honing process parameters for predicting the values of process performance characteristics. A comparative study between ANN and RSM has also been discussed in this chapter. Chapter 7: It contains the conclusions of present research work and future scope for further study.