Keeping your world up and running.®

Building a soil moisture meter and testing it for market

From the Entrepreneur's desk

May 2013

Ever had an idea for a product and thought about taking it to market? Ever started your own company? Steve Sparks became an entrepreneur in 2006 and, after nine months, is in the final stages of developing his first product. It's a low-cost environmentally-based electronic device for determining soil moisture. He's finishing up final functional and reliability testing now with hopes for a first, limited release by early second quarter of 2007.

Sparks began working at Texas Instruments after graduating from high school and six years later completed his Bachelor of Science in Electrical Engineering from Southern Methodist University. Since then, he's worked in both the military and consumer electronics, from ASIC design to software development, hardware and software testing, re-engineering, customer support, applications engineering, and vendor selection and management. And now, he's doing all of it, all at once, on his own.

His big idea

Soil moisture sensors have been around for a number of years but are now (re)gaining interest due to regional droughts/water shortages and environmental conditions, population growth, and the lack of new reservoir construction. Landscapers and other specialists take soil measurements to help guide how much water is needed for optimal plant growth, balanced with water conservation efforts.

The thing is, most people probably don't know how much water they use to water their yard. Sparks' estimations suggest a typical .25 acre lot on an automated irrigation system can use around 2000 gallons of water for each watering cycle. Assuming two watering cycles per week, that's over 15,000 gallons of water/month. In comparison, a family of four uses about 8,400 gallons/month*. Now, if that yard was irrigated for too long, because every soil has a limit to how much water it can accept over a given time interval, then the excess water results in runoff. This wasted water then carries fertilizer and pesticides from landscapes to drinking water reservoirs and rivers. Waste also occurs when automated systems begin watering during or just after rainfall. This wasted water is not only bad for the environment, it can encourage plant diseases in lawns and landscapes.

Right now, soild moisture measurement is accomplished by a number of methods, including electrical (including resistance, capacitance and heat dissipation); monitoring environmental conditions like wind, rainfall, and solar radiation; neutron probes that use radioactive particles; and satellite imagery that employs thermal infrared and/or active microwaves. Sparks had his own idea for a soil moisture measurement sensor.


Steve Sparks hopes that if a low-cost, easy -to-use water conservation device is readily available, more people will reduce their water consumption. His device is composed of basic components such as resistors and capacitors for tuning the design, as well as logic to detect and compute soil moisture, and components to interface to the 24 V ac line. The design fits on a compact PC board and must be encapsulated to work in the outdoors environment.

Soil Moisture probe ->
Soil moisure analysis (Is soil moisture adequate?) ->
Irrigation System Interface (If YES, turn off irrigation; if NO, allow irrigation to run longer)

Although his design's component content is under 20 devices, it uses both analog and digital technology, which adds some complexity. And, since most automated landscape irrigation systems use 24V ac solenoids to turn zones on/off, Sparks' design had to interface with this voltage.


Once his device was operating the way he intended it to, Sparks built several prototypes for more thorough testing. He set up a test bench with two Fluke 189 DMMs, a Fluke 199C ScopeMeter test tool, and an LCR meter. The autoranging and high accuracy of the 189 influenced his decision, given the level of ac and dc current in his device.

Due to the topology, Sparks has to measure both voltage and current throughout his design, which makes having two DMMs a time saver. The 189's "Leads" warning in particular helped reduce the number of times he hooked the leads up incorrectly when changing measurement modes. He also appreciates that fuses can be easily accessed for replacement.

In debug mode, he also uses his tools to check - and record -- resistances, capacitances, diodes, continuity, ac current and voltage, and dc current and voltage. Occasionally, he has to back track and re-measure the capacitance of a capacitor, the forward bias voltage of a diode, the resistance of a resistor, and the continuity of interconnections. That's if he accidently overdrives a device or shorts something out during probing.

As a startup, Sparks says it's critical to keep good records of the tests he performs, for legal and patent reasons. Initially, he used the 199C ScopeMeter test tool for most of his logging, since it has the higher bandwidth and accuracy of the visual signal to get the basic circuit parameters set. As he moves on to environmental testing, he'll switch to the 189 DMM. Then he'll be looking more at the effects of a change rather than capturing the event in time that may have caused the change.

He uses the 199C primarily for looking at signal quality on the AC circuitry, particularly on the solenoid and output driver of the device. Why?

1) Interestingly enough, he can drive the irrigation sprinkler solenoid with a "half-wave" AC signal. A half-wave is an AC signal without the complete positive and negative supply cycle. Although the solenoid will work under these conditions (it's a slow opening/closing solenoid), he's wary about reliability issues. The characteristics obviously cannot be seen using a DMM, (although a trained ear can hear a slight irregular "rattling" noise from the solenoid when this condition exists) and requires the use of a scope.

2) Sparks has to observe and measure the instantaneous current when the solenoid is enabled. The capture period is only a few microseconds - much too fast for a DMM to catch.

3) He's looking for general anomalies on various signals of the device's internal signals. This insures against glitches or noise that could affect operation. Again, these changes in voltage, for instance, might be in the microseconds time period: Too fast for the meter to catch and the eye to detect. However, he doesn't use the 199C ScopeMeter test tool for high accuracy voltage or current measurements, because it's not as accurate as the 189 DMM.

In essence, he uses the 199C to measure current or voltage changes over time and particularly to view short periods of time, and the 189 to measure the steady state (stabilized or not changing) voltages and currents. He has also used the 199C's Flukeview software to capture waveform glitches and email them to a manufacturer expert for his opinion. The ability to communicate like that saved him a great deal of time, and getting the expert opinion greatly increased his confidence in his design.

What's next

Currently, Sparks is building accelerated-life test fixtures. He plans to test his devices in a hot and cold temperature chamber over a period of time, to see how well the device will function in the temperature extremes of possible operating environments. Normal operation may be in 90°F and for 15 minutes two times a week. In accelerated life testing, he'll heat the device to 120°F, simulating a really hot summer - and turn it on and off in 15 minute intervals, 24 hours a day, for a period of time.

At the end of the testing period, any degradation he measures should correspond to how the device will behave in real life, after an equivalent number of years of use by a consumer. He'll be using the 189 Flukeview software to monitor the devices response (voltage of some nodes) under these conditions over time.

Remember - all this will happen by the second quarter of 2007. How will it end? Keep an eye out in home stores near you for new moisture sensing technology to save on your water bills while keeping those roses beautiful.

*Mayer, et al. Residential end uses ofwater. 1999. American Water Works Association Research Foundation