return to home page
return to list of
publications
Aspects of Applied Biology 48, 1997
Optimising Pesticide Applications
A compressed air direct injection crop sprayer
ANDREW LANDERS
formerly Head, Department of Agricultural Engineering,
Harper Adams College, Newport, Shropshire TF10 8NB UK
Summary
A novel concept of a low cost injection system for the direct injection of pesticides, which incorporates a closed pesticide transfer system was devised. The basic concept is that pesticide, supplied in small volume refillable (SVR) containers incorporating a Macro Valve Connection can be pressurized by means of compressed air. The pesticide can be transferred into the mixing chamber of a direct injection sprayer, in a totally closed manner. Altering the air pressure entering the container by using a stepper motor controlled air regulator, will regulate the flow of pesticide entering the mixing chamber.
This paper describes the development of the above concept into a feasible design and presents initial results of an investigation into its operational performance. Tests showed that a long delay occurred when reducing dose rate due to the large volume of stored air in the SVR. The concept of using compressed air for the injection of pesticides has many advantages, although further work needs to be carried out to reduce time delays.
Keywords: direct injection sprayer, pesticide application
Introduction
Direct injection crop sprayers have been developed to aid farmers reduce their pesticide costs, be more environmentally friendly and aid operator safety. However, the present systems on the market are very expensive and most farmers can not justify the added expense of up to £10,000 to the capital cost of the sprayer. Many of the systems available have short comings such as accuracy of application or time delay from the point of injection to the pesticide appearing at the nozzle. The parallel development of returnable containers will further protect the operator and the environment.
The basic principle of direct injection is that the pesticide and water is kept in separate containers on the sprayer. When the sprayer is activated, a metered flow of pesticide is injected into the water stream, via a mixing chamber, which is situated between the main water tank and the pipe work to the nozzles.
Direct injection sprayers have given the farmer the ability to change the pesticide dose rate while travelling across the field, so that varying soil characteristics or weed concentrations within the field can be allowed for. It is also possible to spot treat weeds within a field by switching the pesticide on and off, resulting in less pesticide being used (Marshall & Landers 1990). The maximum response time to gaining the right concentration at the point of injection needs to be carefully considered, Paice, Miller and Bodle (1995) recommend that the time taken to gain the required flow rate at the point of injection should be under 3.5 ±0.5 seconds.
Another advantage of direct injection sprayers is a reduction in cleaning time and rinsate, Landers (1992a) showed that a 1500 litre sprayer fitted with an AgriFutura 2000 direct injection system could be cleaned within 45 seconds with only 7.2 litres of water and 5.2 litres of supray spraynet (pesticide dispersant).
The use of pneumatic pressure as an injection system on a crop sprayer was first developed by IMAG in Holland, originally using propane which was found to be dangerous and it contaminated the pesticide. IMAG subsequently developed the system to use compressed air to apply phenmedipham to sugar beet (Hoenderken 1988). The use of compressed air to inject pesticides into a field sprayer was also developed by Schmidt. Pesticide was kept in a replaceable pesticide tank and flow was altered by adjusting the pressure difference between the water line and the pesticide tank (Schmidt 1983).
Ghate and Phatak (1990) developed a system which pressurises the pesticide container which in turn forces the pesticide into the mixing chamber. However, they also used compressed air to generate the water flow by pressurising the water tank. Water flow was dependent on air pressure and nozzle size, and pesticide flow was dependent on air pressure and a needle valve in a flow meter. This system has been further developed by Ghate & Perry (1994) to give an accurate application of pesticide with respect to ground speed by using stepper motor controlled air regulators to vary the air pressures within the tanks, therefore varying the pesticide flow rate.
Pesticide manufacturers are beginning to supply pesticide in small volume refillable containers (SVR) which have an integrated extractor valve built into the container. The farmer can attach a coupler to the valve on the container and then either using suction or pressure extract the pesticide. The empty (SVR) container does not need to be rinsed, it is returned to the manufacturer to be refilled.
The use of returnable containers will further protect the operator and the environment from the hazards associated with pesticide use (Landers 1992b). An important benefit of the coupling / SVR system is that container rinsing and disposal is eliminated (Perryman 1993).
Materials and Methods
The initial concept had to be refined into a feasible design solution which fulfilled the main design criteria of a pesticide injection system for a crop sprayer, so that calculations and laboratory tests could be carried out (Gamble 1995) and (Wilson 1995). Martinson (1996) developed a data aquisition and control system using a laptop computer fitted with input/output (I/O) cards and an Adventech Genie programme.
The pesticide containers (SVR) and Macro Valve were kindly supplied by AgrEvo UK Ltd. Two twenty litre containers and one 50 litre container were supplied, the containers are pressure tested to a safe working pressure of 120 psi (8.5 bar), however, AgrEvo UK Ltd wanted the containers to be pressurised to a maximum pressure of 1 bar due to the health and safety implications on pressurised vessels.
The mixing chamber design was based around the AgriFutura Dose 2000 system, since this design has a proven track record for effective mixing (Landers 1992c). Due to the low operating pressure requirement from AgrEvo UK Ltd, it was decided to be inject pesticide mimic into the laminar water flow before the mixing chamber. This was done by using a "Y" connection into the water pipe feeding the mixing chamber.
It became apparent that it would be impossible to relate specific air pressures to required flow rates of pesticide, due to the range of properties of the pesticides (viscosity and density) which will be used and the head loss in the pesticide keg. The system will have to incorporate a control system which monitors the pesticide flow rate and adjusts the air pressure accordingly. Another important point to consider is the ability and accuracy of the system to switch the pesticide on and off as it enters the mixing chamber. This point has been a source of failure for many systems which incorporated metering pumps, since the pumps had to be started or stopped, and the resulting time delay leads to over or under applying the required dose rate. A solenoid valve which has a minimum pressure drop across it and can be operated by a d.c. vehicle electric supply was considered ideal for the trials.
From initial trials using a SVR container of water, it was found a small change in air pressure made a large change in flow rate (by increasing the air pressure by 0.10 bar the flow rate was increased by 0.7 l/min). This result indicated that the air regulator needed to be a precision instrument. Another important consideration is the fact that the air regulator needs to be controlled by a stepper motor, so ideally the torque required to turn the stem of the air regulator should be less than 0.5 N/m.
The system is made from all the components detailed above. It was calculated that a pressure of 0.3 bar above the water pressure was needed to allow a flow rate of 4.8 l/min of the pesticide into the mixing chamber. This calculation was based on the pesticide having the same density as water, and does not allow for the different viscosities.
The air source will have to be a steady supply, so that the air regulator is not affected. To achieve this there will have to be a compressor and a suitably sized air receiver to smooth out any pulsations. The compressor should be sized so that it can easily cope with the maximum flow rate of 4.8 l/min, and the maximum pressure of the compressor should be above the maximum air pressure requirement for the system. The compressor could be driven by a tractor p.t.o. or by a 12Vdc motor.
The air receiver and the SVR container could be subjected to the Simple Pressure Vessels (Safety) Regulations 1991. The whole system is governed by the Pressure Systems Regulations 1989.
To evaluate the effectiveness of the proposed pesticide injection system, the components were installed into a laboratory sprayer demonstration unit (Fig.1) Initially all the tests were done with water in the 50 litre SVR container, as it is a safe medium to use. Later trials will use some of the inert test solutions to mimic various types of pesticide and will be used as a comparison to water.
An RS257-149 flowmeter, capable of measuring flows between 0.25 to 6.5 litres/m was chosen.The flowmeter gave an output of pulses with respect to time which were proportional to flow which was then logged every 1/2 second into the datalogger. This output was converted into a flow rate of litres per minute and displayed on the computer screen by using a simple computer programme.
Results and discussion
Test 1 Operation of the solenoid valve
The objective of this test was to evaluate the effectiveness of using the solenoid valve to switch the pesticide flow on and off.
The output readings from the flowmeter were plotted against time, the results can be seen in Fig. 2. The graph shows that a clean and sharp cut off of pesticide occurs within 1 second when the valve is operated. The valve shuts virtually instantaneously (ie 30ms), however, the false flowrate readings are due to the flowmeter having to decelerate from a high flow rate to zero flow rate. When the valve is opened, after having been shut for 2 minutes, the required flow rate is gained within 1/2 second, however, it can be seen for the first few seconds after the valve is opened that the pesticide dose rate is slightly greater than it should be (with a maximum of 5% for the worst case). This could be due to the air regulator hunting during the two minute time period and therefore charging the SVR container to a slightly higher pressure than is required.
The results indicate that the solenoid valve is a very effective way to switch the flow on and off when used in conjunction with the pneumatic system.
Fig. 2. System response to switching the pesticide mimic on /off when applying 4.8 litres/min of pesticide mimic at a pressure of 1 bar |
Test 2 Ability to alter pesticide dose rate
A series of tests evaluated the ability of the injection system to change the pesticide flow rate due to a change in input, such as a boom section being switched off.
The graphs, Figs. 3-6 show the flowmeter readings during the period of time to change the flow rate from its initial value to the required value, against time. From Fig. 3 it can be seen that if the pesticide dose rate is being increased, the time taken to gain the stabilised increase can be as small as 1.5 seconds when increasing the flow from 0.25 to 0.5 litres/min whilst spraying at 1 bar with 45 litres of water in the container.
Fig. 3. Increasing pesticide mimic flow from 0.25 to 0.5 litres/min at 1 bar pressure. 45 litres of water in the container |
Fig. 4. Decreasing the pesticidesticide mimic flow from 0.5 to 0.25 litres/min at 1 bar pressure. 45 litres of water in the container |
Fig. 5. Increasing the pesticide mimic flow from 2.4 to 4.8 litres/min at 1 bar pressure. 15 litres of water in the container |
Fig. 6. Decreasing the pesticide mimic flow from 4.8 to 2.4 litres/min at 1 bar pressure. 15 litres of water in the container |
However, at the other end of the scale it took 12 seconds to gain a stabilised flow of 4.8 litres/min from 2.4 litres/min when spraying at 1 bar with only 15 litres of water in the container, Fig. 5. The quantity of water in the container effects the time delay, tests show that it took 9 seconds longer to complete the same scenario with only 15 litres of water in the container compared with 45 litres of water in the container. This is because if the container is nearly full there is only a small volume of compressed air, so if the flow rate needs to be increased there is only a small volume air to be compressed, therefore it does not take much time. However if there is not much water in the container there is a large volume of compressed air, and it takes slightly longer to compress the larger volume of air.
The size of increase in pesticide dose rate has an effect on the time delay. Only a very small pressure change is needed to go from 0.25 to 0.5 litre/min, however quite a large pressure change is needed to go from 2.4 to 4.8 litres/min.The system spray pressure has a slight effect on the response time. If the system is spraying at 2 bar instead of 1 bar the compressed air supply will be above 2 bar and therefore air will flow into the container more quickly than the air supply at a pressure of 1 bar.
The time taken to reduce the flow rate is a lot longer than the time to increase it. This is because all the swept volume of compressed air in the container is trapped, and the only way of reducing the pressure (ie reduce the flow rate) is to use up some of the pesticide in the container, therefore giving the compressed air more space, and therefore reducing its pressure.
The time taken to reduce the system pressure is also affected by the amount of water in the container, since if there is only a small volume of water in the container there will be a large volume of compressed air to be reduced in pressure and vice - versa.
Acknowledgements
The author wishes to acknowledge the assistance of Sean Gamble, Richard Martinson and Ian Wilson final year B.Eng students at Harper Adams College. AgrEvo UK Ltd kindly supplied the SVR containers.
Conclusions
1. It was established that the initial concept had problems with excessive response times for changing the pesticide dose rates . The addition of a pressure regulator with a high flow rate and fast relieving characteristics could be used to reduce response times.
2. The results are based on injecting water in a laboratory test unit, the response time to change actual pesticide flow rates may be longer, in some cases, due to more viscous solutions being used.
3. The results indicate that the proposed system has a very effective method of switching the pesticide flow on and off; this having been the downfall of many previous injection systems.
4. The concept of using compressed air in conjunction with SVR containers has definite advantages over previous injection systems.
References
Gamble, S. (1995) An investigation into developing a simply, inexpensive and cost effective method of monitoring the performance of a direct injection sprayer. B.Eng investigational project (unpubl.) Harper Adams College
Ghate, S.R. and Phatak, S.C. (1990). A new sprayer for applying pesticides and biocontrol agents. Paper No. 90-1056, ASAE. St.Joseph, MI: American Society of Agricultural Engineers.
Ghate, S.R. and Perry,C.D. (1994) Ground speed control of pesticide application rates in a compressed air direct injection sprayer. Transactions of the ASAE. 37 (1) 33-38
Hoenderken, J.A. (1988). Hoe staat het met de injectie van gewas-beschermingsmiddelen. Landbowmechanisatie 10, 38-39.
Landers, A.J. (1992a). The characteristics and performance of a direct injection crop sprayer. Ph.D thesis. University of Bath
Landers, A.J.(1992b) Closed handling systems for pesticides. Paper given to the Pesticides Application Technology Panel, Nobel House, Smith Square, London. 21st October 1992
Landers, A.J. (1992c). An evaluation of the Dose 2000 direct injection crop sprayer. Proc. of Ag Eng '92 International Conference on Agricultural Engineering, Uppsala. p.336.
Swedish Institute of Agricultural Engineering, Uppsala, Sweden.
Marshall, I. and Landers A.J. (1990). When its better to close for business. Power Farming 79 (9) pp 24-27
Martinson, R.B. (1996). The development of a chemical injection system for a direct injection sprayer. B.Eng investigational project (unpubl.) Harper Adams College
Paice, M.E.R., Miller, P.C.H. and Bodle, J.D. (1995). An experimental sprayer for the spatially selective application of herbicides. Journal of Agric. Engineering Research. 60, 107-116
Perryman, M.A. (1993) The development and use of a coupling system for small volume refillable containers. In: Proc. ANPP-BCPC second international symposium on pesticide application techniques. Strasbourg, France, 22nd-24th September 1993. pp 305-312 Thornton Heath: British Crop Protection Council
Rider, A.R. and Dickey, E.C. (1982). Field evaluation of calibration accuracy for pesticide application equipment. Transactions of the ASAE 25 (2): 258-260.
Schmidt, M. (1983). The direct injection technique for preparing the spray mix - a method of reducing safety and hygiene problems in plant protection. EPPO Bulletin 13 (3) 513-520.
Wilson, I.A. (1995) The development of a chemical injection system for a direct injection sprayer. B.Eng investigational project (unpubl.) Harper Adams College