Bureau of Reclamation Banner

TFLI Water Level Sensor Development

Project ID: 3593
Principal Investigator: Tony Wahl
Research Topic: Agriculture Water Supplies
Funded Fiscal Years: 2006 and 2007
Keywords: None

Research Question

* Can an accurate, economical water level sensor that readily interfaces to Supervisory Control and Data Acquisition (SCADA) systems be developed using load-cell technology?

Need and Benefit

Water level sensors are one of the basic components necessary for successfully improving the effectiveness and efficiency of irrigation water distribution and use. Water level sensors convert actual water levels in canals, flumes, and weirs to electronic signals that can be incorporated into modern SCADA systems. A variety of technologies are available for water level measurement, including pressure sensors, bubblers, acoustic devices, radar, and float-based systems. Each technology has advantages and disadvantages, but no commercial technology has yet emerged with a combination of high sensitivity, long-term stability in extreme environments at low costs.

In the mid-1990s, researchers at the University of Tennessee-Knoxville needed a large number of low-cost, accurate, reliable water level sensors for an anticipated field project. After finding no commercially available sensors that met their needs at a reasonable cost, they developed the Tennessee Fluid Level Indicator (TFLI). The development and subsequent testing of the sensor were documented in a paper published by the American Society of Agricultural Engineers in 1997.

The TFLI proved itself to be very economical, highly stable, and easily adaptable to a variety of water level measurement ranges. Furthermore, the device can be easily constructed primarily from polyvinyl chloride (PVC) pipe components and uses a simple load cell as the sensing element. The many water districts that today build and assemble their own float-based sensors should be well equipped to construct TFLI sensors.

One drawback to the TFLI in its present form is the fact that the load cell sensor is a four-wire bridge circuit and must be driven by a data logger that can provide a precision excitation to it and then measure the resulting output, which is typically less than 30 millivolts (mV). Many data loggers do not have this capability, often due to a lack of sensitivity to measure the output signal. To make it easier to integrate TFLI sensors into existing and new SCADA systems, there is a need to provide an electronic interface to the load cell sensing element. This electronic interface would be driven by a nonprecision 12 or 24 volts of continuous current (direct current) (VDC) power source. The interface would then provide the precision excitation voltage to the load cell and measure the resulting output. The ratio of the output and excitation voltages would indicate the water level, and the interface device would provide the water level as its output in the form of an industry-standard SDI-12 digital signal. This output would be readily integrated into any modern data logger or SCADA system.

A second significant improvement that could be made to the TFLI would be to ease the process of obtaining a reading from it while visiting a field site. A promising approach to doing this is to provide the output over a Bluetooth wireless interface. This interface would allow a visitor to a field site to quickly see the output of the TFLI on a PocketPC or other handheld personal digital assistant (PDA). The Bluetooth interface could also facilitate integration of TFLI data into SCADA systems. Initial development work on a Bluetooth interface has already begun at the University of Tennessee.

Contributing Partners

None

Research Products

Contact the Principal Investigator for information about these documents.

This information was last updated on August 22, 2014
Contact the Research and Development Office with questions or comments about this page