Masters Theses

Date of Award


Degree Type


Degree Name

Master of Science


Biomedical Engineering

Major Professor

Fred D. Tompkins

Committee Members

Luther R. Wilhelm, Robert S. Freeland, Gilbert N. Rhodes Jr


An experimental laboratory-model direct chemical injection system was designed, constructed, and evaluated at The University of Tennessee Department of Agricultural Engineering in Knoxville, Tennessee. Evaluations consisted of determining the transient period between initiation of chemical injection and achievement of full chemical concentration at the nozzle. The laboratory sprayer apparatus was also used to determine variation in chemical concentration in nozzle effluent both from nozzle to nozzle across the boom and with time. Performance using three injection points was evaluated for this system. Points included injection immediately upstream of the system pump, injection immediately downstream of the system pump, and injection at the individual nozzles. Tests were conducted at system operating pressures of 171, 275, and 378 kilopascals. Three injection pumps were also evaluated at the upstream injection point, and two pumps at each pressure-side injection site. The three pumps included one peristaltic pump and two piston pumps. The two piston pumps were used for pressureside injection at both locations. A computer model for predicting transient times for low-pressure injection was also written and validated. Finally, flow characteristics within a conventional application system using a tank mix instead of direct injection were evaluated at the same three pressures to allow comparison with the different injection systems.

The sprayer was equipped with nine flat fan nozzles, and effluent samples were simultaneously taken from each nozzle. A potassium bromide solution formulated at a concentration of 28.3 grams per liter was used as the simulated pesticide to be injected into the diluent stream. Conductivity of the effluent solution caught at the nozzles was measured and related to chemical concentration based upon a calibration of the conductivity meter performed prior to each test.

Results of laboratory studies indicated that performance of the direct injection system was very dependent upon component selection and system configuration. Direct injection systems when used for low-pressure injection with any of the three pumps produced chemical concentrations in the nozzle effluent equal in uniformity to those achieved through conventional tank mixing.

Injection on the high-pressure side of the system pump was effective in reducing the transient period in comparison to injection on the low-pressure side of the system pump. However, mixing of the diluent and the concentrated chemical was reduced. The reduced level of mixing was probably due to the fact that the chemical did not pass through the sprayer system pump, which was found to be effective in thoroughly mixing the two fluids.

When injecting at the individual nozzles, high system operating pressures produced increased variation in chemical concentration in the nozzle effluent. Further, location of diluent entrance to the boom became a critical issue. Flow to both sides of the boom must be equal to achieve uniform chemical concentration from nozzle to nozzle across the boom.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."