Chemical concentration is a vital parameter for determining appropriate chemical application. This study describes the design and testing of a sensor that attempted to monitor concentration of chemicals upstream from each nozzle body. The sensor is based on an LED and photodiode pair. Its ability to detect chemical concentration within the main carrier was tested with a 2,4-D formulation, a glyphosate formulation, and a powdered Acid Blue 9 dye. The liquid herbicide formulations of glyphosate and 2,4-D were tested across common application concentrations of 0% to 12.5% by volume. The powdered dye produced a much stronger effect on the sensor and was only tested at the much lower concentrations of 0 to 50 mg L-1. Further tests were conducted in which the dye was mixed with the herbicide formulation before the combined solution was added to the carrier. While this enabled establishment of pre-determined sensor outputs based on given concentrations of the pre-mixed solution, the sensor may have been responding to the predominance of a dye mixed with a herbicide formulation and not directly to the concentration of the herbicide. While the sensor did not appreciably respond to the concentration of the glyphosate formulation, it did respond in a consistent manner to the 2,4-D formulation and the dye. The sensor‘s response to the concentration of these chemicals was a rational (1/x type) relationship, and the R2 values for the rational models describing these relationships were greater than 0.99. With the mixed dye and herbicide formulation, the effects of the dye and the 2,4-D formulation combined independently, and the total sensor output was a multiplication of the percent effect of each alone. The test with the pre-mixed dye and 2,4-D formulation produced the expected 1 V output at a 12.5% by volume concentration of the 2,4-D formulation, proving that dye can be added to a herbicide to produce a desired response from the sensor. Overall, the sensor‘s response was remarkably stable, with a maximum standard deviation of 42.2 mg L-1 of 2,4-D active ingredient for samples taken at a constant chemical concentration. These tests confirmed that the sensor could respond to chemical formulations and dye in a consistent and predictable manner. However, use of the sensor for herbicide monitoring will require sensor calibration for each combination of herbicide and dye mixture, as the light transmittance properties of the tested mixtures were not quantified and the light transmittance properties of formulations and dyes can be arbitrarily changed by manufacturers.

Document Type


Publication Date


Notes/Citation Information

Published in Transactions of the ASABE, v. 59, issue 5, p. 1089-1099.

© 2016 American Society of Agricultural and Biological Engineers

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

Digital Object Identifier (DOI)


Funding Information

This work was funded with the support of Case-New Holland. This work is supported by the USDA National Institute of Food and Agriculture (NIFA) Hatch Multistate project under 1001110.

Related Content

Publication No. 17-05-084 of the Kentucky Agricultural Experiment Station, published with the approval of the Director.