By Susan Dieterle
Farmers may soon be looking deeper into their grain yields, thanks to an unusual sensor
being developed at Iowa State University.
ISU researchers believe X-ray technology, a proven industrial-inspection technique,
could be effectively used in a grain-yield sensor. And they are encouraged by early
findings that indicate they are on the right track.
"A new sensor using X-ray technology should make farmers' lives easier and should
improve the accuracy of the yield data," said Selcuk Arslan, an ISU graduate student
in agricultural and biosystems engineering. He is collaborating on the project with Feyzi
Inanc, a member of the X-ray group at ISU's Center for Nondestructive Evaluation, and Joe
Gray, who heads the center's X-ray group.
The project is still in its early stages, but the researchers say X-ray sensors appear to
offer a variety of potential advantages over sensor currently on the market. The X-ray
sensor is accurate, insensitive to grain moisture and has a wide dynamic range that
enables it to measure a variety of flow rates without needing recalibration. Also, the
sensor works without coming into direct contact with the grain, so constant cleaning is
unnecessary.
"After looking at the numbers from the lab experiments, we believe that our accuracy
is very competitive with other technology on the commercial market and has fewer drawbacks
than the other approaches," Inanc said.
Yield sensors provide information for farmers and agronomists. The units may be mounted on
combines to measure grain as it is harvested. This helps farm operators pinpoint
problem areas where yield can be optimized by varying the agricultural inputs in those
specific locations.
"It's important to be able to identify the cause and effect of the yield in a
particular part of the field," Arslan said. "Accurate yield maps make the
decision-making process easier for farmers and agronomists."
Yield sensor accuracy is a concern for grain buyers and sellers, and many factors can
affect yield sensor accuracy. Sensors must be accurate within 1 to 2 percent of grain's
actual weight.
One of the most commonly used sensors is an impact-base sensor that measures the force of the grain as it strikes the sensor. The unit then correlates the measured force to grain flow rate.
However, because the grain interacts with the sensor, weed sap, dust and dirt can coat
the device's surface and may reduce accuracy if not properly cleaned.
When grain is wet, the sensor must take the moisture into account to provide a true grain
yield estimate. Moisture changes may affect kernel density, which can change grain impact
characteristics.
Most sensors must be recalibrated every time a different crop is harvested.
"Let's say you're harvesting corn one day, and the next day you harvest soybeans
or move to another cornfield where crop and field conditions change significantly. You
need to recalibrate the sensor," Arslan said. "Obtaining calibration accuracy
can sometimes be very tedious, and some farmers don't always follow all of the calibration
procedures because it takes too long. The recalibration process also interrupts
harvesting."
The changing grain flow rate in the field can also be a factor in accurate sensing. A
combine may encounter everything from bare patches of ground to thickly planted areas so
the sensor must have a wide enough dynamic range to measure changing flow rates without
constant recalibration.
As a Ph.D. student, Arslan explored possibilities for a yield sensor that would be
accurate and easy to use. He considered a gamma-ray sensor like those used in Europe. But
he discovered that U.S. regulations made it difficult to use the radioactive materials. He
then turned his attention toward improving existing sensors and partnered with Inanc and
Gray to study an X-ray method..
X-ray technology has long been a standard assessment tool for nondestructive evaluation,
which involves assessing the integrity of structures and materials without destroying
them. An X-ray yield sensor had never been tested, and Inanc and Gray ran
preliminary tests to see if X-rays might measure grain yields..
Results looked promising, so the team applied for a grant from Ames Laboratory -- a U.S.
Department of Energy research facility on ISU's campus. An Ames program that offers
seed money for novel ideas in need of testing provided $19,500 to build a lab model to
test X-ray technology effectiveness for measuring grain yields.
For the research system, X-ray equipment measures grain passing through a low-energy beam
on its way to a collection bin. Information is relayed to a computer every second and
immediately translated into yield data. Inanc and Gray said radiation from the X-ray is
low enough to allow 2-3 millimeters steel shielding to protect the equipment operator.
No radiation risk is posed to grain or other crops passing through the beam.
Corn was used to test whether the approach would work, but Inanc and Gray believe it will
also work with soybeans, potatoes and other grains.
"Our approach can handle a range of anywhere from bare patches of ground to a flow as high as 20 kilograms per second without recalibration," Gray said. He added that the X-ray sensor is insensitive to moisture.
"With other systems, you'd be measuring the corn plus whatever moisture was in
there," he said.
Developing a demonstration system was relatively simple, the researchers said. Inanc and
Gray modified existing X-ray techniques, and "we didn't buy anything special except
the grain box," Inanc said.
Arslan said the project's next phase -- moving from a laboratory model to a working field
model -- will present some engineering challenges.
"The generator we used in the laboratory model was bulky, so we probably need to
move to a hand-held generator," he said, noting that they may also need to change the
configuration of the system depending on whether it will be mounted on a combine or a
grain elevator.
The researchers have applied for a patent on their technique and seek funding to build a
field model. They say some equipment manufacturers have shown interest.
Inanc and Gray agree that the sensor research has provided them an opportunity to apply
their skills to a different kind of nondestructive evaluation problem.
"It's very interesting work because the problem is from the agricultural field and
the solution is from an industrial-oriented discipline," Inanc said. "It is an
interdisciplinary approach, and we are happy that we came up with some techniques to
address these problems."
About the author: Susan Dieterle is a communications specialist at Ames Laboratory,
111 TASF, Iowa State University, Ames, IA 50011-3020; phone, (515) 294-1405; fax, (515)
294-3226; dieterle@ameslab.gov; http://www.external.ameslab.gov.
Published: March 1999
Last revision: 6/14/99 sd
Home | Comments | Search | Disclaimer