Industries / Food, farming and forestry / Agriculture and farming / Grain elevators and storage
Drying
Drying removes water from harvested grain, usually with heated air, until it is dry enough to store without spoiling: about 15% moisture for corn, by the US Energy Information Administration's figure.
4
Milestones achieved, from 1964 to 2013.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1964
Dry hot, cool slowly
Purdue University engineers first described dryeration in January 1964. Corn leaves the dryer hot, rests ("tempers") for 4 to 10 hours in a separate bin, then is cooled slowly by fans, which removes another 1 to 3 percentage points of moisture. In small laboratory tests reported by Purdue, kernels with stress cracks, fine internal fractures that make corn break in handling, fell from 43.6% with conventional drying and cooling to 7.6%, and breakage from 11.3% to 6.7%; batch dryers gained about 50 to 60% in throughput. University of Minnesota Extension still recommends the method and credits it with 50 to 75% more dryer capacity and 20 to 40% energy savings.
LessonMuch of the cracking comes from cooling hot kernels too fast, so slowing the cooling protects quality and frees the dryer for more grain.
Credit
George H. Foster and Bruce A. McKenzie, Purdue University
1992
A wet crop spoils anyway
The 1992 US corn crop came in unusually wet. In February 1993 Purdue's Grain Quality Task Force reported corn going out of condition in farm and elevator bins across northern Indiana, even for farmers who said they had dried and cooled it well, and elevators charging more discounts for heat and mold damage. Much of it traced to in-bin dryers: in the task force's example, a 24 ft bin needed 41 hours instead of 27 to dry 26% corn instead of 22% corn to 15%, the top layer and the grain by the wall stayed wetter than the average, and many bins were filled too deep.
LessonA dryer can reach the right average moisture and still leave wet layers that spoil in storage.
Credit
Dirk E. Maier, Purdue University
2005
Husk-fired dryers fill the Mekong
The flat-bed dryer, which blows heated air up through a layer of grain, was originally developed by the International Rice Research Institute (IRRI), but its kerosene burner made it costly to run, and it did not take off in most countries. Nong Lam University in Vietnam enlarged it to 4 to 8 tonnes a batch and fired it with rice husks, a waste product of milling. By 2005 about 4,000 of these dryers were installed in the Mekong Delta, by IRRI's count, and in 2019 the CGIAR rice research program counted about 10,000 in Vietnam, 1,700 in Cambodia, 1,000 in the Philippines and 400 in Myanmar.
LessonA dryer spread once it ran on a free local fuel and was sized for the groups that used it.
Credit
Nong Lam University; IRRI
2013
Drying strains propane supply
A wet growing season and the largest corn yield in US history, by the Energy Information Administration's (EIA) account, sent propane demand for crop drying to a record for November 2013: US use reached nearly 1.8 million barrels a day in the week ending 1 November. Midwest propane stocks fell 12.8 million barrels from mid-October to January, against a five-year average fall of 7.3 million. With early cold weather, a pipeline outage and rail delays, prices spiked in January 2014, and several states suspended limits on propane truck drivers' hours.
LessonDrying a large wet crop can draw so much fuel at once that it leaves too little for winter heating.
Credit
measured and reported by the US Energy Information Administration
Watch for
Heat pumps dry Ontario grain
Engineers at the University of Guelph are building a full-scale prototype batch dryer that dries grain at low temperature with electric air-source heat pumps in place of fossil-fuel heaters. Their model, checked against three experimental tests, says that adjusting compressor power and fan speed every hour, based on the outside air, gives 7.86% better efficiency than adjusting the compressor once a day.
Achieved whenthe full-scale dryer dries grain on a farm through a harvest, and its energy use per tonne of water removed and its running cost against a propane dryer are measured by someone outside the team.
Claimed by
Joel Csajaghy, William Lubitz and Mostafa Sharqawy, University of Guelph, May 2026
Open
01
Stress cracks counted by phone
A free phone app that photographs 100 corn kernels laid germ-down on a light pad and reports the share with no, one, two or many stress cracks, the four counts Purdue uses to judge drying damage.
The briefCredit · unclaimed02
Humidity cards, scored
A side-by-side test of the humidity checks small farmers can afford to tell whether maize or rice is dry enough to store, the DryCard humidity card and a cheap digital hygrometer sealed in a jar with a grain sample, against a laboratory water activity meter and the oven, publishing how often each check calls wet grain dry.
The briefCredit · unclaimed
Open 01 · the brief
Stress cracks counted by phone
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A free phone app that photographs 100 corn kernels laid germ-down on a light pad and reports the share with no, one, two or many stress cracks, the four counts Purdue uses to judge drying damage.
Why it is still open
Purdue's fact sheet tells producers to inspect 50 to 100 kernels by eye on a light board. Li and colleagues (2022) built an industrial prototype that finds stress cracks with 92.7% precision and 94.4% recall, but not a tool a farm or elevator can download. The nearest phone app, from the University of Wisconsin–Madison, scores how well a forage harvester has cracked silage corn, not drying damage. Web searches in November 2026 found no stress-crack app.
Built with what exists
A phone camera; an LED tracing light pad; a 3D-printed tray that holds 100 kernels germ-down; Label Studio for labelling; an Ultralytics YOLO segmentation model; Purdue's four crack categories.
First step
Print a 100-kernel tray, photograph 10 trays of corn from one dryer, and have two people count each tray.
Done looks like
200 trays (20,000 kernels) from at least five dryers or dryer settings, each counted by two people; the app's stress-cracked share within 5 percentage points of the two people's average on at least 90% of trays, with the two people's own disagreement reported beside it; images, labels and code public.
Open 02 · the brief
Humidity cards, scored
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A side-by-side test of the humidity checks small farmers can afford to tell whether maize or rice is dry enough to store, the DryCard humidity card and a cheap digital hygrometer sealed in a jar with a grain sample, against a laboratory water activity meter and the oven, publishing how often each check calls wet grain dry.
Why it is still open
Air sealed with grain settles at a humidity set by the grain's moisture, so a humidity reading can stand in for a moisture test. Bradford, Dahal and Bello (2016) showed that humidity indicator paper can estimate seed and grain moisture this way. The DryCard's developers say users in more than 30 countries found it effective and that it has been "proven effective", but the abstract describes no trial. The nearest field test, of one low-cost capacitance meter on maize in Benin markets, found it agreed with the oven on which side of an 18% storage line a sample fell only 61% of the time, and did not test humidity checks. A search of Crossref and Europe PMC found no published rate at which the DryCard calls wet grain dry.
Built with what exists
DryCards (about US$1.50 each, by their developers); airtight jars; a calibrated digital humidity sensor, such as Sensirion's SHT45 on a breakout board; a benchtop water activity meter, such as METER Group's AquaLab, as the reference; a laboratory oven and balance; Python.
First step
Buy 20 DryCards and one humidity sensor, bring 30 maize samples to known moistures by adding measured water, and read them with both checks.
Done looks like
300 maize and rice samples between about 10% and 20% moisture, read by both checks at 20 and 30 °C; for each check, the share of samples with a water activity above 0.65 (air above 65% relative humidity, the level below which the DryCard's developers say fungi stop growing) that the check calls dry.