Equipment Limitations and Challenges in Precision N Management

1
Equipment Limitations and Challenges in Precision
N Management

• R.K. Taylor, G. Dilawari, P. Bennur, J.B. Solie,
  N. Wang, P. Weckler, and W.R. Raun

2
Variable Rate Liquid Applicators

• Direct Injection
• Complex
• Time lag
• Fixed orifice nozzles
• Flow proportional to square root of pressure
• Difficult to achieve range in flow rates
• Pulse Width Modulation
• Variable orifice nozzles
• Pressure/Flow relationship is more complex

3
Objective

• To determine the consistency among nozzles and
  repeatability of individual nozzles with respect
  to flow rate for commercially available variable
  orifice nozzles

4
Materials

• Nozzles Eight of each type
• TurboDrop Nozzles (GreenLeaf technologies)
  TDVR-02 and TDVR-03
• VeriTarget Nozzles (SprayTarget )

5
Test Stand

• Pump
• Wet boom equipped with TeeJet 3 nozzle bodies
• Pressure relief valve
• Throttling valve

6
Methodology

• Three nozzles were selected randomly and tested
  on the three nozzle boom at 20, 30, 40, 50, 60
  and 80 psi
• 8 repetitions for each pressure
• Flow was adjusted to the three nozzles to achieve
  the desired pressure.
• Outflow from nozzles was calculated by measuring
  the volume of water collected from each nozzle
  over a period of 30 seconds.

7
Data analysis

• Analysis of Variance was performed in SAS 9.1
  (SAS, Cary, NC) using PROC ANOVA to detect flow
  differences among nozzles.
• Data for each nozzle were analyzed by pressure
• Means were separated using the LSD option and at
  0.01 level of significance
• Pressure-flow curves for each nozzle type were
  plotted and compared with the manufacturers
  pressure-flow data.

8
Results TDVR-02
9
Results TDVR-02
10
Results TDVR-03
11
Results TDVR-03
12
Results VeriTarget
13
Results VeriTarget
14
Conclusions

• Inconsistent behavior was observed between the
  nozzles at different pressures
• Repeatability of a nozzle was better at pressures
  above 40 psi.
• Both the TurboDrop nozzles performed according to
  manufacturers specification
• CV for VeriTarget nozzles, for most of the
  nozzles, was around 10 which is acceptable for
  spraying

15
Sensor v. Map Based VRA

• In a map based system, the controller receives a
  rate change as the applicator crosses into a new
  zone.
• However, with a sensor based system the
  controller typically receives an updated rate
  every second and does not have the opportunity to
  stabilize.

16
Sensor Configurations
17
Accepted Resolution
18
Materials

• Raven 440 controller
• Raven Fast Close valve
• Data acquisition with flow meter and pressure
  transducers

19
Test Stand Schematic
20
Model
1st order valve response Proportional Integral
controller
21
Input Data
22
Prescribed Rate
23
Model Output
24
1 second lag
25
Output Data
26
Conclusions

• The modeled results showed a mean absolute
  application error of 12.9 L ha-1.
• These results further indicate that the predicted
  response lagged the prescribed rate by
  approximately 1 second.
• This resulting misapplication could be reduced by
  half if the controller delay was reduced by 1
  second.

27
Addendum
28
Improved System ??
29
Questions
Randy Taylor Randy.Taylor_at_okstate.edu 405-744-5425