Optimization of Design Parameters and Performance of a Portable Common Beans (Phaseolus Vulgaris L) Thresher

Abstract

In Kenya, threshing of common beans at smallholder level is mainly by traditional methods using sticks and animal trampling which are slow, inefficient and tedious. Combine harvesters are suitable for overcoming such difficulties. However, the existing economic structure of agricultural production coupled with the small size of farms in Kenya makes the use of combine harvesters uneconomical. Portable threshers offer an alternative solution to threshing of beans for smallholder farmers, but they are rare in the local market. Therefore, this study focussed on development of a portable common beans thresher that is customized to the needs of small-scale famers (working up to 10 ha). The design process was through modelling, simulation, optimization and fabrication of the thresher. The mathematical models of the machine were developed by dimensional analysis through the concept of Buckingham pi theorem and reference to other similar work in literature. The objective function for optimization involved minimizing grain damage and power required for threshing while maximizing threshing efficiency and throughput capacity of the machine using Taguchi method. The results from simulation and experiments show that there were strong to very strong correlations between simulated and actual data as the coefficients of determination were greater than 0.7. The differences between the means of the simulated and actual data were not statistically significant at 5% level of significance for power requirement and throughput capacity. A prediction performance of 77% was attained when actual and simulated data were compared at 10% absolute residual error. It was observed that power requirement, bean grain damage, threshing efficiency and throughput capacity increased with increase in pegs peripheral speed from 1.88 m/s to 5.65 m/s for actual and simulated data. The optimum mean values for bean grain damage of 1.73%, threshing efficiency of 99%, throughput capacity of 57.95 kg/hr and power requirement of 556.6 W were obtained at 3.765 m/s peripheral speed of pegs, 0.023 kg/s feed rate, 1 m linear concave width, 17.5% moisture content (w.b) and 5.5 kg weight of threshing cylinder. Threshing efficiency of 99% was achieved for the developed bean thresher for all cylinder rotational speeds greater than 600 rpm. The thresher attained throughput capacities of 72 and 125 kg/hr at feed rates of 1 and 2.5 kg/min, respectively. Mechanical bean grain damage was less than 3.3% for cylinder rotational speeds of 600-1000 rpm. In addition, the beans thresher consumed one litre of petrol to thresh 30 kg of beans. Finally, the thresher had a cleaning efficiency of 78.7% at optimum cylinder rotational speed of 800 rpm. The developed common beans thresher may require improvement on cleaning efficiency and automation, otherwise it is recommended for use by the small-scale bean farmers.