Date of Graduation

5-2018

Document Type

Thesis

Degree Name

Master of Science in Biological Engineering (MS)

Degree Level

Graduate

Department

Biological and Agricultural Engineering

Advisor/Mentor

Christopher G. Henry

Committee Member

Thomas Costello

Second Committee Member

Christopher Bobda

Third Committee Member

James Randall Andress

Keywords

Android app to microcontroller, Bluetooth-low-energy, Flowmeter, Paddlewheel, pump-timer, supercapacitor

Abstract

For better irrigation efficiency, it is recommended that farmers track their water consumption to avoid over-irrigating. However, it is difficult to implement this as it is labor intensive to supervise pumps manually and available technologies require high investment. Therefore, a rain sensing pump controller for 3-phase electric irrigation pumps and a stand-alone portable Android enabled paddlewheel flowmeter has been developed to test their feasibility.

The pump timer is a retrofit device for irrigation pump panels. The controller allows an irrigator to start and stop the pump with less supervision. An infrared rain sensor is integrated with the controller to measure rainfall and terminate pumping at a preset precipitation threshold. Also, the pump controller has no batteries and automatically restarts the pump if pumping is interrupted by a power outage. Functional testing of 30 days demonstrated zero false rainfall interrupt during actual rainfalls and an average accuracy of 10% in rainfall depth measurement.

The paddlewheel flowmeter uses a microcontroller integrated with a recently developed communication protocol known as Bluetooth Low Energy (BLE API). The flowmeter calculates and transmits real-time flow data to an Android app. However, due to component limitations, the flowmeter lacks storage memory, and therefore the totalizer reading is reset to zero when power is interrupted. The prototype uses an optimized paddlewheel design. Nine different 3D-printed plastic paddlewheels were tested to improve paddle efficiency. Three distinct blade designs, each printed in 3 blade-count variations of 4, 6 and 8 blades were compared to a turbine meter. The diameter of the axial shaft was also reduced to minimize drag. The most efficient paddlewheel design was determined by its regression slope, startup speed, and relative flow-measurement errors. The average error of the conventional steel paddlewheel was reduced from 17% to +-1%. However, to estimated flowrate the final software uses the Kc-factor which was found to be 18,700 pulses/acre-inch at all flowrates. The flowmeter utilizes solar power to charge the supercapacitor bank. The completed prototype was operated for 108.5 hours to test functionality and resulted in an error of 1% in flowrate and a difference of 5.5% in total volume.

Download

Share

COinS