La dérive de pulvérisation peut être mesurée au champ ou en soufflerie. Paradoxalement, les tests au champ sont sujets à de grandes variations dues aux conditions atmosphériques mais peuvent être plus facilement réalisés contrairement aux tests en soufflerie. Ainsi les principaux modèles de dérive sont basés sur des mesures au champ alors que peu de modèles s’inspirent de mesures en soufflerie. L’objectif de ce travail a été de définir un ou des descripteurs de la dérive sur la base de l’analyse de courbes de dépôts obtenues dans la soufflerie d’IRSTEA Montpellier. Par rapport aux souffleries existantes, un protocole d’exposition de longue durée a été utilisé avec une forte densité d’échantillonnage. Un plan expérimental comprenant 99 modalités a été réalisé en tenant compte de différents types de buses (FF, AI, AI Twin jet), différente hauteurs de rampe de 40 à 80cm, différentes positions de la rampe (frontale, latérale et angles intermédiaires) et différentes vitesses de vent entre 2 et 7.5m s-1. Les résultats ont montré que le taux de dérive à 5m sous le vent (DR5) correspond au descripteur le plus robuste si l’on tient compte du large spectre de paramètres et de réglages. Des modèles de premier ordre ont été définis pour l’expression de l’effet de la vitesse du vent ainsi que de la hauteur de rampe selon le temps de vol des gouttes (ToF). Ainsi il est possible de comparer des résultats issus de conditions expérimentales différentes et de simuler l’effet de la vitesse du vent et la hauteur de la rampe pour un type donné de buse. Des mesures in situ de taille de gouttes ont confirmé la pertinence du temps de vol comme base de l’expression des résultats. / Spray drift might be measured either infield or in a wind tunnel through specific sampling strategies. Paradoxically field tests are subjected to a high variability due to the atmospheric conditions but can be more easily conducted in the absence of a wind tunnel. The result is that most of spray drift models are based on infield measurements. Conversely very few models were developed on the basis of wind tunnel measurements. The objective of this work was to define spray drift descriptors from the analysis of drift curves in IRSTEA wind tunnel. Compared to the majority of existing wind tunnels, a long duration exposure protocol was applied with a high sampling density. A large experimental plan of 99 modalities were conducted including nozzle types (FF, AI, AI Twin jet), boom heights from 40 to 80cm, boom positions (frontal, lateral, and intermediate angles) and wind velocities from 2 to 7.5ms-1. Results showed that the drift ratio at 5m (DR5) was the most robust drift indicator considering the wide range of parameters and operations conditions (wind velocity, boom height). First order models were drawn for the expression of the effect of the wind velocity and the boom height according to the droplet time of flight (ToF). As a result it was possible to compare data from different experimental conditions and to simulate the effect of the wind velocity and the boom height for a given type of nozzle. In situ droplet size measurements confirmed the relevance of the time of flight expression.

Master of Science / Department of Biological & Agricultural Engineering / Robert E. Wolf / Traditionally, growers spray uniform application of pesticides over the target area regardless of variations in pest infestations. In recent years, variable rate application (VRA) technologies have made it possible to apply pesticides in variable rates across the field. In pesticide application, nozzles play a vital role. In general, pesticides are applied using conventional nozzles. Most conventional nozzles vary flow rates only over a 2:1 range when operated within the recommended pressure range due to a fixed spray orifice. Conventional nozzles vary droplet sizes tremendously when there are speed and application rate changes which results in inefficient application. Conventional nozzles have limitations when used for VRA. A new nozzle called Varitarget nozzle (U.S. Patent No. 5,134,961) was developed and marketed by Bui, (2005) to overcome the limitations with conventional nozzles. Varitarget nozzles have a variable orifice that changes in size in response to pressure changes, allowing varying flow rates with a minimal change in droplet size. Laboratory tests and field tests were conducted to study the performance of Varitarget nozzle. Varitarget black/blue and clear/yellow caps were evaluated in this study. Lab studies were conducted to measure Varitarget characteristics compared to conventional nozzles. The flow rate ratios of Varitarget nozzle black and clear caps were 12:1 and 10:1 while the conventional nozzles produced flow rate ratios ranging from only 3:1 to 4:1. The measured flow rate of Varitarget nozzle black and clear caps was similar to that published by the manufacturer up to 40 psi and varied higher after 40 psi. Both Varitarget black and clear cap nozzle was within the standard VMD requirements until 40 psi and showed increasing trend while the conventional nozzles matched the standard VMD requirements. The VT black and clear cap nozzles showed better coverage at higher pressures when compared to conventional nozzles. CV values for VT black and clear capped nozzles were less than 10% which indicates capability of good uniform distribution. Spray angle of 110 degrees for VT black and clear capped nozzles was consistent over a range of pressures. Field studies were also conducted to compare the Varitarget to conventional nozzles. In the varying speed study, droplet size varied from 498 to 621 microns with a SD of 47.50 for VT black nozzle and 465 to 599 microns with a SD of 54.08 for VT clear cap nozzle as the speed varied from 4 to 12 mph. In the varying application rate study, The droplet size varied from 432 to 510 microns with a SD of 27.84 for VT black nozzle and 355 to 452 microns with a SD of 39.80 as the application rate varied from 4 to 12 GPA. In both studies, the observed pressure range required for spraying was minimum and varied slightly.