The seed that wouldn't fit

Watch first

The machine wants a seed. The farmer has a rhizome.

Ginger comes out of the ground small, long, thick, curved, irregular. Same crop, five shapes. A planter built around a fixed passage meets that variety and does what fixed passages do: the wide pieces will not enter, the curved ones rotate sideways and jam, and the whole thing stops.

So the farmer sorts first, then plants the rest by hand.

The problem was not planting ginger. It was expecting ginger to be uniform.

Then I stopped trying to measure it

The instinct is to grade the seed: build sizes, sort into them, feed the machine what it wants. That moves the work onto the farmer and throws away good seed.

The better question was whether the tool could simply adapt to the piece. Instead of different holes, one adaptable opening. Small rhizome, fingers close. Large rhizome, fingers open.

Don't measure the ginger. Let the tool adapt to it.

Design principle: make the mechanism forgiving.

The solution

Rhizo-Drop is a planting pocket that changes shape around whatever is put into it.

• A soil-opening shoe that cuts the furrow
• A self-adjusting rhizome pocket, gravity fed
• Flexible compliant side guides that close or open to the shape and hold the piece at its centre
• A release gate that drops it cleanly into the soil
• A soil-closing shoe that covers gently, plus field wheels and a handle

One mechanism, different sizes. No perfect size required.

What changed

The old sequence was sort, select, fit, place, cover, with a rejection pile at the start. The new one is load, adapt, plant.

Good mechanisms do not fight variation. They absorb it.

The farmer did not need uniform seeds. The tool needed to stop expecting them.

What I ruled out

Grading the ginger first. Sort it into sizes, then plant each size with matched settings. It works, and it adds a whole job before the job. The point was to remove handling, not move it earlier.

One pocket big enough for the largest rhizome. Simple, and it wastes spacing and seed on every piece smaller than the biggest, which is nearly all of them.

A vibrating singulator. The mechanical answer, and it is what jams. A mechanism that separates by shaking assumes the thing being shaken has a predictable shape, which is the assumption that started the problem.

What already exists

Potato planters handle irregular tubers, and they are the closest prior art. They use cups sized to a graded range, which means they depend on the grading step I was trying to remove. Precision seeders are the other end: excellent, and built for seed that is uniform by definition.

Ginger sits between the two, and the machinery market mostly does not serve the middle.

What I have not tested

The jam rate, which is the number this design lives or dies on. Also untested: whether spacing stays accurate when the pocket is adapting constantly, how wet rhizomes behave against dry, and whether the mechanism survives a season of grit.

What I would do next

Count misses per hundred. Skips and doubles are the only honest measure of a planter. A forgiving mechanism that skips one seed in ten is worse than a rigid one that needs grading.

Test with the worst batch, not an average one. A sample sorted by eye before testing would prove nothing, because the whole claim is about what happens when nothing has been sorted.