ABSTRACT

Mathematical models for predicting the performance of cocoyam chipping machine were developed by dimensional analysis using the concept of Buckingham's Pi theorem. The models were developed through selection of various chipping parameters and operating conditions affecting cocoyam chipping process. Two cocoyam varieties: NXs 001(ede ocha) and NXs 002 (ede uhie) were used for the study and their physical properties were determined and compared. Measured properties include: moisture •content, size (length, width and thickness), geometric mean diameter, sphericity, weight, surface area, bulk density, true density, porosity and angle of repose. Results obtained showed that the cocoyam variety NXs 001 has higher average moisture content of 57.8 1% while the variety NXs 002 has the average value of 46.97%. The average length, width and thickness of the varieties NXs 001 and NXs 002 were (10.06cm, 4.74cm, 4.16cm) and (9.97cm, 3.66cm, 3.09cm) respectively. The variety NXs 001 has higher average values in geometric mean diameter 3.80cm, sphericity 0.80, mass 127.66g, surface area 45.37cm2and porosity 52.4%, than the NXs 002 with values in geometric mean diameter 2.90cm, sphericity 0.79, mass 81 .22g, surface area 26.42cm2. and porosity 50.96%. A comparative evaluation of the two varieties of cocoyam for the above properties revealed that only the differences in the means of length, sphericity, weight and true density were not statistically significant at the 5% level, while the differences in the means of the other properties were all found to be statistically significant at 5% level. The average values of the data obtained from the two varieties of cocoyam were used for the study. The models developed include: Chipping efficiency, throughput capacity, power requirement, chip damage and loss models. The developed models were optimized by a minimization technique. Computer simulation visual basic program at three factor interactions of the process variables was used to select combinations of optimum crop and machine variables such as cutting velocity, chipping slot size and feed rate that will yield minimum chip loss for cocoyam chipper. The results obtained from the simulation revealed that minimum chip loss was obtained at 0.0018m2chipping slot area, 0.056kg/s feed rate and 4.9lmls cutting velocity. The optimum parameters obtained were approximately 0.05kg/s throughput capacity, 0.17kg total chip loss, 961 .40Watt power requirement, 0.01 kg/s damage rate and 80% chipping efficiency. These optimum values were used to design and develop the cocoyam chipping machine. The results of the machine performance test showed that the highest output capacities and efficiencies of 196.12kg/hr and 85.18% were obtained respectively at the speed of 375rpm. Results of the performance evaluation of the cocoyam chipper obtained were subjected to further statistical analysis using a 3x5 factorial experiment in Randomized complete Block Design (RCBD). The results of the analysis show that cutting velocity within the range of 3.6 — 4.9lmls had highly significant effect at P0.0 ion the throughput capacity, power requirement and total chip loss, but had significant effect only at P0.05 on the chipping efficiency of the cocoyam chipper. The result also show that the chipping slot area within the range of 0.0014 — 0.0018m2, had highly significant effect at P 0.0lon throughput capacity, chipping efficiency and total chip loss, but had no significant effect at P0.05 on the power requirement of the cocoyam chipper. The developed models were verified and validated with established experimental data from the developed cocoyam chipper. All the models developed were found to conelate and fit well with experimental data with Rsquare values greater than 0.90 at 0.05 level of significance.