Inventors Notes – Switching Electronic Air Pressure Regulation

My Asparagus harvester invention utilizes pneumatic cylinders to cut the individual spears, and the stroke length has to right on the money every time. If the pressure goes up the stroke length becomes longer, and if the pressure goes down the stroke length shortens.

Too much pressure and the piston rod will bottom out against the front cylinder head, and not enough pressure will reduce the stroke length an cause the blade to not cut all the way through the spear or even not reaching the spear at all. Allowing the piston to bottom out against the front head will eventually damage the cylinder.

The asparagus harvester has 14 air cylinders mounted on the header arranged across the asparagus bed. Each piston rod is equipped with a sharp blade with a slight bit of overlap with the blades next to it. The cylinders are angled down toward the ground and when they extend the blade severs the spear slightly below ground level requiring a stroke length of about 20 inches. Typically the extension stroke takes around 35 to 40 milliseconds.

An optical detection system locates the spears and sends a signal to open the air valve for the cylinder corresponding to the co-ordinates of the spear to be cut. The harvester is moving forward at between 20 and 30 inches per second, and so the blades must cut the spear and get back up out of the way of any spears that are not quite tall enough to harvest.

Asparagus spears emerge from the bed in a random pattern with random heights. At any moment during harvesting there may be as many as 5 or 6 cylinders operating at the same time, or none at all. You might have 10 feet with nary a spear, and 18 spears in the next 24 inches.

Because these cylinders are very fast acting they require high flow rates at a constant stable air pressure. While stroking, the cylinder will be consuming around 165 cubic feet per minute of air. Six cylinders operating at once would require a whopping 990 cubic feet per minute.

With such large swings in flow and rapidly varying air consumption the mechanical air regulator will have a significant variation in the pressure drop, which will have a detrimental affect on the stroke length of the cylinders.

We can, however, use another approach to regulating the air pressure. We can use a switching electronic air pressure regulation scheme. With this approach we replace the mechanical pressure regulator with an on or off electric air valve with a high flow rate.

We can then use an accurate analog pressure transducer to open the valve whenever the pressure drops below the set point, and shut off when the pressure is at or above the set point.

The valve has a very low pressure drop unlike the mechanical regulator. The valve can handle the flow required by multiple cylinders without the air pressure drooping that the mechanical regulators end up with.

There will be small pressure spikes or what is known in electronics as a ripple in the pressure. By properly sizing the manifold I can filter out the small pressure ripples.

For more details about electronic switching air pressure regluation for the asparagus harvester

Inventing a Selective Asparagus Harvester

For the last week or so I have been trying to do some life cycle testing on the pneumatic cylinders that we are going to use for the next asparagus harvesting machine we build. The cylinders should be able to do about a million strokes before the need to be replaced.

The machine has a row of pneumatic cylinders, or often referred to as air cylinders, arrayed across the asparagus bed. As the machine moves forward a sensing system locates the spears and tells the air cylinder lined up with the spear when to cut.

The cut signal causes the piston rod to extend from the cylinder at high speed with about 18 inches of stroke. It takes less than a tenth of a second for the cylinder to extend to full stroke. On the end of the piston rod is mounted a blade that cuts the spear.

I set up a fixture out in my garage for testing the cylinder. I’ve got it mounted to a frame similar to how it will be mounted on the asparagus harvester, pointed down at the ground at around 45 degrees.

I filled an asparagus crate or lug box, lined with plastic, full of dirt from the back yard. I placed the crate of dirt so that when the cylinder is extended the blade goes about two inches deep into the soil.

I actually went to the grocery store and bought a bunch of asparagus to test the cutting ability of my blades. I wanted to see if I could detect a difference between a blade with a V notch in it, a slanted edge like a guillotine, and an arrowhead type blade.

I tamped the soil down till it was nice and firm, and then used a dowel to make a hole just big enough to get an asparagus spear into. Then I pushed a spear into the hole and tamped the dirt down around it. I lined up three spears so the blade would contact the first spear while in mid-air, the second spear right at ground level, and the third spear would have the cut line about an inch below ground.

I tried this with all three blade types, and I could find no difference at all in the cutting ability or anything else. The blades sliced through all three spears like they were made of butter. There was no deflection or twisting of the blade, so my new secret method of preventing blade rotation seems to work well.

Since revealing details about an invention online would compromise my patent rights I can’t go into details about the new method I am using to prevent the blades from rotating out of position.

I would like to do the life testing at 150 psi, but my compressor only goes between 120 psi and 135psi as it cycles. So I set the air pressure for the testing at 120 psi.

I’m interested in the life of the seals, and whether the piston rod ends up breaking due to metal fatigue. The load placed on the end of the piston rod by the blade and guiding assembly is offset from the center of the piston rod.

On the down stroke the pneumatic valve reverses the direction of the air to the cylinder before the cylinder reaches the physical end of its stroke to prevent damaging the cylinder. On the return stroke the piston hits the rod end of a smaller cylinder screwed into the rear head of the cutting cylinder to act as a spring and absorb the shock loads.

My compressor can just barely keep up with the cylinder if I fire the cylinder every 20 seconds. It’s going to take a long time to get anywhere near a million strokes. I need a much bigger compressor.

To cycle the cylinder I used a 12f675 micro controller chip, an 8 pin chip with a microprocessor, memory, and various interface modules like analog to digital converters, comparators, and counters all included. Even an accurate clock is built in. Learning to program and use these microcontroller chips should be in every inventor’s toolbox.

I programmed the chip using a basic language. I used a breadboard, a couple of pots and a voltage regulator etc along with the chip to create an automatic cycling controller. It has two pots. One pot controls the time between firings and the other determines the length of the pulse sent to the air valve. The longer the pulse the longer the stroke produced by the air cylinder.

I’ve tested a whole lot of air cylinders with this method and I’ve yet to find one that would even go 10,000 cycles without developing a problem. I think this time I’ve got an air cylinder that will hold up for that million strokes I need.

These new cylinders I’m using have a 1” diameter bore. The cylinders I’ve used previous had a 1-1/2 inch bore. There is a big difference. The smaller surface area of the piston means the force is much smaller. The acceleration is determined by the force, and the new cylinder is much more sensitive to variations in pressure. That is something that the asparagus harvester invention will have to address.

In a future article I will describe in some detail the pressure problems and the special electronic air pressure regulation system I intend to use for the machine.

To learn more about my selective asparagus harvester invention visit: Selective Asparagus Harvester

Inventing an Asparagus Harvester – 30 Years of Prototypes

I first decided to invent a selective asparagus harvester in about 1972. I asked my dad one day if he knew of something that needed inventing… I was bored. He was a farmer and an asparagus grower. He told me they needed a mechanical asparagus harvester that would just harvest the ripe spears; a selective asparagus harvester.

Not knowing any better I decided I would build one. At the time I was about 21 years old, fresh out of the army, and pretty good with electronics. I had a ham radio license when I was about 13 years old, built my own transmitters and receivers, and could fix anything from a car radio to a color TV.

I had read about a new kind of imaging device, I think it was one of the first CCD chips. I don’t remember all of the details, but the camera had basically 16 rows and 16 columns of light sensitive elements, and I decided to use that to detect the height and location of the spears on the bed, and I would use blades attached to air cylinders aimed toward the ground at about a 45 degree angle. I used eight cylinders arranged in a row across the bed, and when the camera spotted a spear tall enough to cut, it would activate the valve and fire the air cylinder that was lined up with the same column as the spear.

A friend of mine and I built a little demonstration prototype that had a little gas powered air compressor built out of channel and angle iron and 4 motorcycle tires that we pushed by hand. It had the camera, air compressor, 4 cutting cylinders and a crude pickup device that would grip the spears as they were cut.

That first prototype was enough to interest a local machine shop that decided to take risk of developing a selective asparagus harvester. They hired me for $2000 a month to oversee the development and took a 50 percent share of the rights to the machine. We spent the next ten years working on it, coming up with a new prototype each year.

The camera turned out to be unsuitable for the task, and during those years I tried just about everything you could think of to detect those stubborn spears of asparagus. I tried little wire bales that hung down from above, beam-breaking photo electric sensors, retro-reflective optical sensors, magnetic switches with plastic paddles, and even a Reticon line scan CCD camera, but all had serious drawbacks.

I really wanted to try a laser for illuminating the spears due to the precise position information I could get by using a laser shooting across the bed. It would be able to give me much more accurate information about where the spear was located on the bed and it’s height. But at the time lasers were several thousand dollars, and not nearly rugged enough to mount on an asparagus harvester.

Asparagus spears can be very delicate, and on cold mornings it is very easy to break a spear by just nudging it a bit. So you really don’t want to use something that has to contact the spear to detect it. Using through-beam sensors required mounting the emitter and receiver at the height of the spear you wanted to harvest. If you wanted to cut nine inch spears you mounted the beams nine inches above the bed. Harvestable spears would range from nine inches to about 16 inches on hot days. The longer spears fortunately are harder to break.

We used extremely thin sensors to avoid touching the spears, but you could still see the occasional spear break as it made contact with the sensor itself.
Another problem with sensing the spears was the fact that asparagus spears can lean in any direction, and significantly throw off the targeting of the spear. At the point where the spear reaches the nine inches off of the bed, it can be several inches to one side or to the front or back of where the spear actually emerges from the ground. That makes it a whole lot harder to cut the spear. Especially if the blades are narrow.

In 1984 we gave up the project due to lack of interest on the part of the asparagus growers. The machine was a self-propelled 3 row selective asparagus harvester. It wasn’t perfect yet but it did harvest asparagus.

At that time we used beam-breaking for sensing the spears; I think they were 4-1/2 inch wide channels the spears had to pass through.

The sensing of the spears and locating them were not the only problems we had in developing a selective asparagus harvester.

My next blog entry will discuss the difficulties we had with the air cylinders. And some of the inventive ways we found to address the problems… and why most of them did not work.

The Old Inventor Guy