Search

A direct injection system that injects pesticides locally at all nozzles on a boom was assembled and investigated under laboratory conditions with regards to its ability to be cleaned. The system was contaminated with a safe-to-use polyvinylpyrrolidone-water solution as a test pesticide before cleaning. The cleaning process was divided into two steps: 1) reclaiming the simulated pesticide by pushing it back into the pesticide tank using pressurised air (pre-cleaning) and 2) rinsing the contaminated part of the hydraulic system with water. Cleaning performance was investigated systematically to ascertain the initial dynamics of the rinsing process as well as long term cleaning. Evaluations included variation of 1) pre-cleaning time and 2) water inlet position. Measurements showed that the concentration of the simulated pesticide in the rinsing water could initially be as high as 30%. As the pre-cleaning time was extended the initial concentration was reduced by one third. Changing the water inlet position reduced the initial concentration of the simulated pesticide in the rinsing water to 5%. These concentrations were much higher than in most common spray solutions. This means that if an active pesticide was used further dilution of the pesticide concentration in the rinsing water would be required for it to be sprayed on a crop. In some cases it took more than 10 min to dilute simulated pesticide residues in a 3 m test section down to 0% which clearly shows the need for more research on factors influencing the cleaning process and more development of cleaning processes.