Direct injection systems for agricultural spray applications continue to present challenges in terms of commercialization and adoption by end users. Such systems have typically suffered from lag time and mixing uniformity issues, which have outweighed the potential benefits of keeping chemical and carrier separate or reducing improper tank-mixed concentration by eliminating operator measurements. The proposed system sought to combine high-pressure direct nozzle injection with an automated variable-flow nozzle to improve chemical mixing and response times. The specific objectives were to: (1) integrate a high-pressure direct nozzle injection system with variable-flow carrier control into a prototype for testing, (2) assess the chemical metering accuracy and proper mixing at different combinations of injection valve frequency and duty cycle along with chemical pressure, and (3) assess the ability of the control system to ensure proper chemical dilutions and concentrations in the nozzle effluent resulting from step changes in target application rates. Laboratory experiments were conducted using the combined system. Results of these experiments showed that the open-loop control of the injectors could provide a means of accurately metering the chemical concentrate into the carrier stream. Chemical injection rates could be achieved with an average error of 5.4% compared to the target rates. Injection at higher duty cycles resulted in less error in the chemical concentration predictions. Discrete Fourier transform analysis showed that the injection frequency was noticeable in the nozzle effluent when the injector was operated at 3.04 MPa and 5 Hz (particularly at lower duty cycles). Increasing the injection pressure and operating frequency to 5.87 MPa and 7 Hz, respectively, improved mixing, as the injection frequency component was no longer noticed in the effluent samples. The variable-flow nozzle was able to maintain appropriate carrier flow rates to achieve product label chemical concentrations. In one case, the maximum allowable concentrate was exceeded, although the nozzle was able to recover in 0.5 s. Steady-state errors ranged from 2.5% to 7.5% for chemical concentrations compared to the selected chemical to carrier ratio (0.03614). This test scenario represented an application rate of 4.68 L ha-1 with velocity increases from 4.0 to 7.1 m s-1 and decreases from 7.1 to 4.0 m s-1, which were typical of the example field application data.

Document Type


Publication Date


Notes/Citation Information

Published in Transactions of the ASABE, v. 62, issue 2.

@ 2019 American Society of Agricultural and Biological Engineers

The copyright holder has granted the permission for posting the article here.

The document available for download is the authors' post-peer-review final draft of the article.

Digital Object Identifier (DOI)


Funding Information

This material is based upon work supported by the Cooperative State Research, Education and Extension Service, U.S. Department of Agriculture, under Agreement No. 2009-0720-1146.