I've had the idea for a donkey-powered mill for awhile, and recently I came across a good automotive differential to be the primary hub and gearcase for the project. With typical abandon I went to work and had the basic machine running about 10 days later.
Sebastian was the test pilot for this project. I hope to have three sweeps eventually and use the whole team. A threshing machine will require all the power I can get. This sweep is a carbon-fiber windsurfer mast, much stronger than it looks in the picture. I might go for something a bit longer eventually. The skid is just spiked into the ground right now. The sheller is just sitting on the ground, too.
Here is how the differential looked the day I brought it to the farm. The seller told me it was from a rear-wheel drive Mazda, but I have no way of confirming that. It took some doing to get the old driveshaft and rear axle off.
I liked this differential because all three shafts terminated in nice 4-bolt flanges. Easy for fabricating.
One of the axles needs to be prevented from turning. This differential offered an easy solution--I added a piece of angle iron to serve as both mounting bracket and the stop-turn mechanism. It got a little trickier, however, when I found that any pressure on the axle flange caused the gears inside to bind up. I later learned that the seal around the axle is not very good, and oil leaks out. If I get the unit inside again, I think I will clean it up and put some good silicone around the lower axle to keep the oil in better.
So I fabricated another mount to hold the unit up around the driveshaft flange. Since nearly every differential has a modest gear reduction from the driveshaft to the axle, it makes sense to drive the donkey mill from one of the axles. In theory there would be no problem attaching the sweep arms to the driveshaft input. You would just lose some rpm on the output shaft. By driving one of the axles I'm getting about a 1.5:1 speed increase. My donkeys pulling on 9' sweeps are doing about 3 rpm I figure.
Here I am using some bicycle sprockets to get an additional 43:13 gear ratio. I used 1/2" shaft for my interim driveshaft, but now I wish I had used at least 3/4". It looks like a potential area of weakness.
The long shaft coming out of the center is 1/2" water pipe. Again, maybe I should have gone bigger. I fabricated some simple flex couplings from pipe flanges, little pieces of rubber cow mats, and 1/4-20 bolts. I put these flex joints at both ends of the 1/2" pipe driveshaft.
Here is the unit to receive the power coming off of the sweep. Again, very quick and dirty. I'd like to improve this with some kind of shut-off clutch and some additional shafts to make additional gear ratios easy to achieve. I'd like to run a fanning mill and a small threshing machine eventually, in addition to the corn sheller. The corn sheller made gearing pretty easy for the first time out, since it's made to run with a hand crank at around 20-30 rpm I think. The final output speed tested was 30-32 rpm, with Cassie on the sweep. Other items will require higher rpm and more gearing. I felt good about using a v-belt for the final drive, since the slippage would provide some safety as I built up the system and learned how it worked.
After running about 100 lbs of corn through the mill, the 1/2" pipe shaft seems a little undersized. It would buck under loads, usually when the donkey would slow down and the sheller get stuck with an ear of corn half way through. I'd like to rebuild it so the shaft has a higher rpm, and therefore lower torque on it. I think a second chain on the center before transferring power to the main shaft is a good idea. Also, I'm wondering if I can find some sort of slip clutch to act a safety overload.
I resolved over the summer that I needed to start working on a threshing machine. The grain coming off the homestead plot is getting bigger every year, and hand threshing is a pain. The wheat was especially a problem, and I left a lot of seed in the straw.
My threshing machine has to conserve seed and prevent cross contamination. These are both nearly impossible if I use the combine as a threshing machine. The combine can almost never be clean, and small batches of threshing will be very wasteful. Due to the size and complexity of even a small combine (or old threshing machine), there is probably 10 lbs of seed sitting inside it at any time. The test plot may only produce five or ten pounds of one variety of wheat in one season, so the scale is just not suitable.
I looked around for something to give me a leg up in building a threshing machine, but could not find anything. I had resolved to build one from the ground up, using these plans as a departure point. I thought I could at least get some old rasp bars from a dead combine somewhere, and that's where I found my leg up.
It turns out that nearly every Massey combine built has separate threshing drum for tailings, which they call a re-thresher. While most combines (and old threshing machines) route un-threshed heads back into the main threshing drums, Massey thought it would be better to make a small, axial-flow head to deal with this stream. The re-thresher is a little strange. It feeds from the side and discharges at the top. I think this makes it technically an axial-flow drum. It's a bit like a combination squirrel cage fan and threshing drum.
Anyhow, these re-threshers are very well built and it had a lot of what I wanted to get the project going. There are good rasp bars and solid bearings. The concave is adjustable by adding or removing spacers under the concave bars. Although I may try to make a true concave and change the flow to ordinary threshing drum flow, for the time being I am experimenting with it as is.
Here is what the concave bars look like. I experimented with sunflowers, wheat, and beans. For beans I removed all the bars. There are spacers under the bars to change the concave clearance. The beans were still getting cracked with the bars removed and the rpm reduced to 406 rpm. Slower rpm might help.
So there is no separator right now, just a tarp. The re-thresher throws the beans pretty far.
I threshed 25 lbs of beans in two sessions. It went pretty quickly.
I built a V-plow a couple winters ago for the team to clear work paths. I can keep work paths open between the house and barn now very easily. It clears a lot of snow very quickly, although there is not too much control. It can get a little squirrely, and once there is an established snow bank it's hard to change the path. But you can't beat it for quickly opening up a path.
Work photos are below. Sorry if I'm a little hazy about some of the details.
The plow under construction. I think those are 2x10's. I have a small gait that I want to get through, so this plow is not particularly large. It cuts a path around 6' wide. The angle is 90 degrees. In discussion with others I learned that a "pointier" design might possibly make the plow track better.
I made a snath for a new scythe blade that a friend gave me for my birthday. I had planned on making it out of a solid piece of ash, but the tree I thought was an ash turned out to be an elm. The wood was too far gone so I had to change plans.
The completed snath and blade. I finished the wood in ordinary spar varnish. There are actually 21 pieces of wood total!
I remembered a piece of clear Douglas Fir sitting in the basement rafters, so I decided to use that for the project. I already had a good scythe that I liked, so I used it as a pattern. The fir was cut to length and then ripped on the table saw thin enough to make the required bend. I made a jig on the workbench, glued things up, and clampled the pieces together. I think the main part of the snath was 6 pieces thick.
I began by making the straight piece, with a bend towards the bottom. I used my workbench for the straight part, and clamped on a jig for the bend at the bottom. It's a mild bend, less than 15 degrees.
For the lower handle I had to rip the pieces thinner, to make a sharper bend. I did this in a separate lay-up. I had to glue on extra small pieces to make the snath thicker in the area of the handles, and where the blade attaches. I used attachment hardware from an english-style scythe.
I cleaned up the glue joints with a block plane, then went back to the workbench to jig up the lower handle. I had to rip these pieces a bit thinner to get the bend I wanted. The handle is glued onto the main part of the snath without any nails or screws.
The handles are some clear wood from the elm tree that I had hoped to make the snath out of originally. I shaped the handles and then put a 7/8" diameter on the end. The snath had 7/8" holes drilled in at the appropriate locations, and the handles were glued in place. The entire snath is held together with ordinary carpenter's glue.
Here is what it looked like after a good shaping and sanding. I think it is plenty strong but only time will tell. It's a little fancy for something that is supposed to be workmanlike, but I enjoyed this project.
Attachment hardware is from an ordinary English-style scythe that was rotting away in my collection. The Tops blade from Scythe Supply was said to have a tang that works well with English-style scythes, and I only had to slightly touch it with a grinder to make it fit. I like having this hardware since the Austrian-style clamp can slip so easily.
The winter wheat on Spring Street is finally looking like a crop! Thank goodness. I saw a few green blades on the 18th, but these pics from the 21st are the first day that the wheat was up uniformly. That makes it a little more than a month since planting on September 18.
The view from the Southwest corner. It's a two acre planting this year.
Spring Street on October 21. The patch on the right is Vavilov, and on the left is Wisconsin No. 2.
Wisconsin No. 2
Vavilov Turkey Red
Ehmke Turkey
Had a great two days plowing in the garden (10/20 & 10/21). We are plowing up the corn ground at the East end and the squash/fallow area on the West end. The West side was not fallow on purpose--it was so wet I could never get in there this year. As a result it's a weedy sod and it's not turning over very well. I hope I can get it worked up. My plan is to make it a tiny hay field.
The plow and the new (third try) disengage mechanism worked very well.
The germination comparison that I planted on October 7 is up well enough to get a good comparison now. The first bits could be seen two days ago.
Wisconsin No. 2. This strain has the most pronounced red color.
The Vavilov Turkey Red still has the red stem, but it's less brilliant.
The Ehmke Turkey does not have any red color, and the leaf blades are wider.
The Stephens is taking a little longer to germinate and does not have any red color, either.
The Spring Street field is still not up! It was planted on September 18, almost a month ago. There has been rain now so it should be coming up in a matter of days.
The backup plot of Wisconsin No. 2 germinated overnight. I have been keeping this ground moist with irrigation since planting it on October 1. Six days to germinate seems a bit long to me, and I'm beginning to think that Wisconsin No. 2 and Vavilov both take longer to germinate than modern "Turkeys."
I'm also intrigued by the red stems that are showing in the WN2. I've never noticed it before. From observing the regular test plot, I can see that the stem does go away after a few days. But it's very apparent today.
I want to get a good comparison, so I drilled in a four-row test plot just to the South of the backup WN2 plot. From North to South, there is a single 8' row of Wisconsin No. 2, Vavilov Turkey Red, Ehmke Turkey, and Stephens Turkey.
The Wisconsin No. 2 on October 7, after six days in warm, moist soil. I've never noticed the red stems before. I expect it will disappear after a few days, but now I want to know if other wheat shows this color in the first days of germination.
Here's a shot a day later, on October 8, around 7 am. I think the red appears out of the ground when the wheat plant shoots up overnight, but is then covered over by the green of photosynthesis after a day or two in the sun. The color comes from anthocyanin, the same substance that makes leaves turn color in the Fall. The color is there throughout the growing season, but it only can be seen shortly before harvest when the culms cease photosynthesizing. At least that is how I understand it!
This is the same spot on October 10. Now the color has been almost covered up by chlorophyll.
Here is the germination test plot, just South of the backup WN2 plot. This is former basil ground that I worked up with a wheel hoe this morning. From North to South, WN2, VTR, ETR, and STR.
Here's the germination test plot in a wider view. Rows run East-West.
Here's a view of the main test plot right now, October 8.
This year was my most ambitious corn planting to date. I grew two varieties of sweet corn, planted four full rows of popcorn, and also had two plots of composite flint corn. I probably had 75 distinct varieties this year!
Here is the composite flint corn from Bill Davison at the University of Illinois. I believe he said there were 50 varieties represented in the colored seed stock. He also sent me a smaller sample labeled "white composite." The white was planted last and it's still growing. The colored flint is spectacular. I have no idea yet what I will use for seed.
Cassie on one of the first days harvesting popcorn. I was disappointed with the yield--there were a lot of stalks with nothing on them.
My seed stock. I look for full ears on stalks that are strong and still standing. If there is any mold or rot I definitely don't use it for seed.
Cassie working the flint corn. Some ears were 7' in the air--wow! Cassie did pretty well. The new cart is not as handy as the old one in tight spaces.
