The successful application of crop processing on pull-type forage harvesters requires engineering analysis and experimentation in areas such as roll design, particle aerodynamic, vibration, power measurement and crop physical characterization. The first step of this research project was to design, build and evaluate two processing rolls integrated in an experimental pull-type forage harvester. With a 12.7 mm theoretical length of cut setting, the processing rolls were used to treat alfalfa at a 1 mm clearance between the rolls. Samples analyzed at an animal research centre showed that processing increased the effective ruminal degradability of corn by 3% but did not improve the degradability of alfalfa. / The presence of processing rolls in the forage harvester increased friction. The air inlet area had to be increased to improve particle flow and obtain maximum air outlet speed. The presence of crop processing rolls required an increase in blower speed of 10 to 20% to maintain the throwing capacity. / Critical speeds were analyzed by the Transfer Matrix Component Mode Syntheses (TMCMS) method and three classical approaches using analytical models to predict deflections in the shaft. Experiments showed that the TMCMS method was very reliable and predicted the critical speed with an error of about 4% compared to 8% or higher for classical methods. Modal analysis also proved to be a simple and practical way to measure critical speed. / A laboratory setup was built to investigate various mechanical adjustments: roll clearances of 4 and 6 mm for corn, and 1 and 3 mm for grass; eight peripheral speed ratios between 1.08 and 1.93; two levels of throughput (9 and 18 t fresh crop/h). Optimal adjustment will ultimately depend on animal response to various levels of crop breakage. / A hydrodynamic model of chopped forage processed between rolls was developed to determine crop properties using an experimental database and subsequently predict the power requirement as a function of the configuration. Forage specific area was estimated using an experimental micro-screening method. A program was developed to calculate mechanical stresses within each roll of the crop processor as a function of working conditions and steel properties. (Abstract shortened by UMI.)

No description available.

A whole-stalk harvester was developed as part of a sweet sorghum-for-ethanol production system. Gathering chains grasped the stalks as they were cut at the base with a disk-cutter. These stalks were flipped onto a cross conveyor and deposited into an accumulator. Periodically the machine stopped and the accumulator was dumped. All the components on the harvester are powered hydraulically. Five pumps on the harvester supply flow to seven actuator circuits. Power is delivered to the pumps from the tractor PTO via a universal joint driveline. Each of the six existing circuits and one proposed circuit were modeled with bond graphs and implemented for computer analysis using TUTSIM. Model validation was done by comparing simulated and measured driveline torque, line pressure, and return line flow rate in each of the six existing circuits. Data collected on the gathering chains circuit was used to analyze the effect of driveline joint angles on transmitted torque and pump output. Torque measurements at three driveline angles showed a torsional vibration with a primary harmonic at driveline rpm and a secondary at twice driveline rpm. A combination of Cardan joint characteristics, mass unbalance, the secondary couple, and non-linear driveline and V-belt stiffness was used to model the driveline. Resulting simulated torque emulated the experimental very well. Measured pressure in the gathering chains circuit showed relatively low fluctuations at the highest amplitude torsional vibration (highest driveline joint angles). It was concluded that driveline vibration would not significantly affect the gathering chains circuit performance. The cross-conveyor motor circuit simulation showed close agreement to experimental results. Mean predicted flow, pressure, and torque were within 8.9, 7.3, and 7.7 percent of mean measured values. A simulation with a stalk load on the conveyor showed that power requirement increased only 8.0 percent. The accumulator dump circuit was analyzed to determine if the load on the motor would become over-running and cavitate the pump or motor as the stalks were being dumped. Simulation showed that a bundle up to 300 kg could be dumped without over-running the motor, and since this was a larger bundle than the bin could hold, a design modification was not necessary. The disk-cutter circuit was designed based on simulation results for several combinations of motor, pump, and sheave ratio. A 7.3 cm³/rad motor, 2.53 cm³/rad pump , and 2:1 sheave ratio produced the correct disk-cutter speed, and low torsional vibration when cutting the stalks, consequently this combination was selected for the design. / Ph. D.