This concept is based on the simple and elegant manual corn sheller. The millet is fed into a hollow bearing where it is beaten, and everything comes out the end of the tube.

The first prototype consisted of a bumpy rubber material in a tube on a drill (video).

The second prototype consisted of two concentric metal pipes as a hollow bearing, one clamped in a vice and the other spun manually using a bike intertube. At the end of the pipe was a piece of PVC pipe with a few staples protruding into it to grab the millet. This video shows a second pass of threshing a millet stalk (most of the grain came off on the first pass). Given the promising performance, we created a more sophisticated prototype using a bicycle to achieve higher rotational speed without changing directions. This bike was donated from Bikes Not Bombs.



The tube must be spun at high speed because the radius of the threshing surface is small, about 1/2 inch. Our measurements from other concept devices revealed efficient threshing at 300 rpm against a 4" diameter wheel. Here the threshing is against ~1" (inner) diameter, so ~1200 rpm will be required to achieve the same speed at the millet seed of ~1-2m/sec.



The inner 1.5" pipe contacts the bike tire and rolls inside a 2" pvc "bearing". WD-40 between the lower parts of the tube reduce friction. The tire is inflated to only ~15psi and only moderate pressure is applied to keep the bearing from binding.

I wide variety of threshing teeth, wires, brushes, etc. were tried including the following:



These were also spun at a variety of speeds. None achieved robust, efficient threshing. Invariably at high speeds the millet stalk begins to rotate with the tube and shears off.



We realized this tendency to rotate and shear is because the teeth or wires apply a rotational torque to the millet, which is one of its weakest axes. Analyzing the force diagrams led us to conclude that this method is fundamentally flawed as a millet thresher. Too bad, as it offered the potential to thresh all sides of the panicle at once, with a single simple motion, as well as constrain all the grain and fiber within a compact space to facilitate winnowing and collection of grain.
Instead, we continue to pursue the rotary drum thresher and the bike spoke thresher (both on left side of force diagram).


Return to Prototypes.