Performance Evaluation of a Pollution Free Draper Separator for Decorticated Groundnut

Abstract

Low cost groundnut decorticators used by small and marginal farmers rarely include a cleaning unit, so the kernel shell mixture they discharge is separated by hand winnowing or by an air blast, both of which are labour intensive and a persistent source of dust. An inclined draper separator, which exploits the difference in shape and surface texture between kernel and shell, offers a pollution-free mechanical alternative, and the present study quantifies the relative importance of its operating settings for groundnut. A manually operated draper separator fitted with a velvet belt was evaluated in a 3 x 3 x 2 factorial experiment laid out in a randomised block design with three replications, comprising 54 cleaning trials: deck angle at 17, 22 and 27 degrees, conveyor speed at 0.19, 0.34 and 0.55 m/s, and feed rate at 45 and 85 kg/h. Cleaning efficiency and shell loss were computed from the recorded masses of clean seed, kernel lost to the shell outlet, separated shell and shell retained in the clean seed; because these four fractions account for the whole of the material fed, every trial is closed by a mass balance, and closure was verified numerically for all 54 trials as an independent check on the recorded data. Cleaning efficiency ranged from 50.4 to 98.0 % and shell loss from 2.7 to 10.9 %. Deck angle emerged as the controlling parameter, contributing 82.7 % of the total sum of squares (F = 418.7, p < 0.001) and raising mean cleaning efficiency from 56.20 % at 17 degrees to 76.39 % at 22 degrees and 93.88 % at 27 degrees, while conveyor speed and feed rate contributed 4.6 % and 5.0 % respectively, and none of the two-factor interactions in cleaning efficiency were significant. The highest cleaning efficiency, 98.0 %, was obtained at a deck angle of 27 degrees with a conveyor speed of 0.55 m/s and a feed rate of 45 kg/h. Shell loss followed a different structure, in which deck angle, feed rate, and the conveyor speed x feed rate interaction were of comparable magnitude, so that it cannot be predicted from main effects alone. Each additional degree of deck inclination was worth approximately 3.8 percentage points of cleaning efficiency, which provides a direct and readily applied design rule. A high deck angle is therefore recommended unconditionally, while conveyor speed should be selected for belt cleanliness and comfortable manual operation, and feed rate should be matched to the capacity of the decorticator supplying the separator. The setting that balances both responses is identified in a companion paper.