Evaluating factors affecting pellet durability and energy consumption in a pilot feed mill and comparing methods for evaluating pellet durability

Doctor of PhilosophyDepartment of Grain Science and IndustryKeith C. BehnkeA series of experiments was conducted to compare methods used to evaluate the durability of animal feed pellets, as well as to investigate the potential for modeling the effects of formulation and processing factors on both pellet durability index (PDI) and pelleting energy consumption, measured in kilowatt hours per ton (kWh/ton). Seven different factors, including ground corn particle size, added fat level, inclusion of distillers dried grains with solubles (DDGS), feed rate, steam conditioning temperature, conditioner retention time, and pellet die thickness (L:D ratio) were examined. Each factor was evaluated at two levels, and treatments were developed in order that all factor to factor comparisons could be made. Pellet samples were analyzed according to the standard method as described in ASAE S269.4, a modification of this method, and by using the NHP100 pellet tester set to each of its four testing intervals (30, 60, 90, and 120 seconds). The standard method was found to provide the most consistent and repeatable determinations of pellet durability, and was found to correlate well with the modified method, as well as with the NHP100 results at 30 and 60 seconds. Physical attributes of feed pellets, such as pellet hardness, bulk density, and moisture content were found to have significant, but weak correlations with pellet quality. Pellet quality was found to be significantly influenced by all factors other than ground corn particle size and feed rate. Higher fat level, lower conditioning temperature, and the thinner pellet die most significantly lowered pellet quality, with increasing effect respectively. A regression model was developed that was able to predict pellet durability within an average of 1.1 PDI. Pelleting energy consumption was found to be significantly influenced by all seven factors, with the higher fat level, thinner pellet die, and higher conditioning temperature most improving efficiency, with increasing effect respectively. A regression model was developed that was able to predict energy consumption within an average of 0.3 kWh/ton. The successful creation of regression equations demonstrates that there is potential for modeling and optimizing pellet quality and energy consumption within a pelleting operation