Modern agriculture uses techniques and technologies that have provided farmers with increased yield and a possible reduction in costs. Optimizing the use of inputs by applying exact and accurate doses, which match the real needs of the soil, in addition to supplying the necessary nutrients for the correct development of the crops, enables a reduction in costs and environmental impacts caused by the incorrect use of products such as fertilizers and pesticides. With this background, this paper presents a study on the development of a capacitive sensor to identify the absence, presence or variations in the distribution of solid mineral fertilizers. To evaluate this sensor, eight different formulations were tested in distribution analysis with an overflow dosing mechanism, both statically and dynamically, with 2% maximum moisture variation between all samples. The identification of an absence or presence of fertilizers was successful in 100% of the experiments. Tests to identify variations in the fertilizer distribution were carried out through simulated obstruction. The sensor identified a reduction in the fertilizer flow in all experiments, obtaining numeric variations above 55%. In the fertilizer formulation identification test, only the formulations 02-28-20 and 06-21-12 in experiments carried out with the overflow dosing mechanism did not differ statistically one from another, while all other formulations presented a statistically significant difference in the ANOVA analysis and the Tukey test at 5% significance.
Keywords: capacitive sensor; dielectric permittivity; dosing mechanism; fertilizer distribution.