PTO
I purchased a used hydro fan to use as the end of the belt driven PTO. I wanted to make sure that if I somehow screwed up building the new PTO that I'd still be able to use the tractor. Having bought the fan, I'll probably never need it, but if I didn't buy it, I probably would. This is my corollary to Murphy's law.

I machined a piece of 2" round steel to fit the shape of the fan end. I then drilled and tapped the fan end to accept a 5/16" - 18 allen bolt. This would not only locate the 2" round on the end of the fan, but also add a little strength to the assembly after the extra piece was welded to the end.  After trying different amperage settings on the arc welder on some similarly sized scrap pieces, I found that I couldn't come to a happy medium in regard to amperage settings. Too low and I could weld the 1/8" fan blades, but didn't get good penetration on the 2" round. Too high and I got good penetration on the heavy stock, but burned through the 1/8". Rather than chance ruining the fan, I decided to braze it where I have better control of how much and where I focus the heat. The job doesn't look that pretty, but the weld seems to take as much punishment as I could easily give it without bending the blades - and that was without the allen bolt installed. With the bolt installed it should be able to handle the rotational stress - or at least I hope so. We'll see how it holds up over time.

Trying out the hydro fan with extension welded on.	An internal square was cut in the extension to match the square milled in the 3/4" bar.

I did a test fit of the bracket so I could mark where to cut the top of the bracket.	I set the bracket back up on the mill, located the hole and bored the hole out to 1.7" with a fly cutter.

There's now plenty of room for the PTO shaft and collar that will set the depth of the shaft in the hydro fan. The cut-out is for the right rear hydraulic coupler	Some drill rod is chucked up in the lathe to make the collar bushing.

The diameter of the chucked end is just under 1.5" - diameter is not critical - and the other end has been turned to 1.24".  This side needs to be close.	Everything fits as it should. I still need to add a set screw to the large end of the collar so it can be locked in place.

Houston, we have a problem. I needed to do a little cutting to gain some clearance around the bracket	Studs are tack welded in place on the back of the mounting strap.

Business side of the stud bracket. Two 1.25" studs 3" on center.	I used a sheet metal screw to hold the stud bracket in place. Simple and effective.

The stud bracket is installed.	The finished PTO bracket along with my fabrication notes.	The CAD drawing in JPG format. Click image for full size or here for the AutoCAD DWG file.

The PTO is about finished.  All that remains is to make a double sheave to power the tiller.	The inner bushing/collar screw sets the depth of the PTO shaft in the hydro fan.

Pulley/Sheave
My Canadian friend, Kevin, supplied me with the dimensions for the double sheave that powers the tiller. I looked around for a 3 5/8" OD double pulley for 1/2" belts for a few weeks or so without finding one that was suitable for this application.  Deere lists the price for this pulley at about $80 and I thought that this was a bit much, so I decided to make my own. All I needed to do was turn two half inch grooves to fit the shape of the 1/2" wide x 5/16" deep belts and space them 3/16" apart.  How hard could that be?

As I have never machined a sheave from scratch before, I did a little research on sheaves and pulleys. It turns out the even though the v-belt included angle is 40°, when machining a drive pulley, the angle of the sides of the groove should be machined  to a lesser included angle.  This is done so that the wedging action as the belt is stretched causes more gripping force between the pulley and the belt. Drive pulleys are generally smaller than driven pulleys and the machined angle differs depending on their use. It is said that 32° to 34° for drive pulleys and 36° to 38° for driven pulleys are the optimum angles for "A" type or 4L belts.

The depth of the groove is also cut deeper than the depth of the belt. The depth of the groove should be deep enough that even when the belt is stretched or worn it will not contact the bottom of the groove.  I will machine the depth to about 0.45".

Although aluminum is not the ideal material for a pulley, I chose it because I have a good assortment of aluminum rounds.  That said, I had an aluminum fan pulley on an old Volkswagen that didn't visibly wear for many years and this pulley will only see use a couple times a year.

The closest chunk of aluminum round I had was 4.5" diameter by 2.25" thick. Almost perfect on the thickness. The double sheaves and space between them worked out to be about 1.3" and I wanted somewhere between 3/4" -1" length for the reduced diameter "snout" that would attach the pulley to the shaft with a couple of set screws.

I started by chucking up the aluminum and facing the one side. Now the width was 2.2". I now needed to turn the diameter of the snout from 4.5" to 2". This would result in a lot of wasted aluminum, so I decided to cut a ring out of the chunk of round. No telling when a ring with an OD of 4.5" and an ID of 2+" would be useful for another project. I used a parting tool to cut a slot in the round and then came in from the face side to separate the ring from the chunk.

Using a parting tool to cut in 2.5". It's a deep cut on a small lathe, so I advanced the bit slowly.	After grinding a thin bit, I cut in from the side until the two cuts met. Results: one ring of aluminum saved for a future project.

The pulley was drilled for the 3/4" shaft and then I tried something that I had only read about - making a key-way slot on a lathe.  I used a very sharp parting tool and with the lathe off, I used the carriage crank to advance the tool into the hole. Each pass was only removed a thousandth or so of material. Very slow going and there was a fair amount of deflection. This meant that the key-way was deeper toward the entry side than the rear. I would even out the depth when I finished cutting the pulley grooves as I didn't want to un-chuck the pulley until all of the cuts were made.

With the round of aluminum now chucked by the snout, I started reducing the diameter.	I measured and marked the lines I would use for reference. These match the pulley on the tiller.

I next cut a relief on the rear side of the pulley. This would allow me to move the pulley a bit closer toward the PTO bearing to set the belt alignment. In hind-sight, a bearing with a slightly shorter inner race would have not required the relief to be cut in the pulley. On the plus side, the closer the pulley is to the bearing, the less leverage force that will be exerted on it.

To cut the groves, I measured and marked the lines I would cut to. I ground a bit with each side angled to 16°. I started in the center of each area I had marked for the groove and by advancing the bit 10 thousandths for each pass, I cut the grooves. I finished the job with some crocus cloth to round over the corners of each groove so that the belt would not get cut.

Using a bit ground to 16° on each side, I started cutting the grooves.	First groove done and starting on the second.

When I was done with the grooves, I removed the pulley from the lathe and used a file to even out the key-way.

The next step was to cut a key-way in the PTO shaft. This was done with an end mill on the milling machine. I then cut a key from some mild steel stock and spent the rest of the evening filing and adjusting until the pulley, key and shaft all fit together nicely.

The grooves have been cut and the edges have been smoothed with some crocus cloth.	Test fitting the pulley on the PTO shaft. Everything aligns as it should. Looks like I'll need some new belts as these don't seem to be a matched pair.

I had a couple choices for holding the pulley on the shaft. I could drill and tap the end of the shaft for bolt that would run parallel to the shaft or use set screws at a right angle to the shaft. Due to not having a shoulder on the shaft to clamp to, I decided on using set screws. I drilled and tapped the pulley for two 1/4"-20 x 3/4" set screws over the top of the key and two more offset 90°.  Probably a bit of over-kill, but I'm betting that the pulley doesn't come loose from the shaft.

Last thing on the PTO agenda was to give the PTO bracket a coat of gloss black paint.  That done, I assembled the PTO and bolted it in place. I have to say that installing this PTO is as easy as installing any other of the Deere accessories I have.

Two pair of set screws. Two over the key and two at ninety degrees. Over-kill, but what the heck.	Drawing for PTO double sheave. Click image for larger view or here for DWG file.

October 12, 2012
I put the gas tank back on tonight and fired up the 317. With the idler pulley disengaged, the tiller tines turn slowly. Unfortunately this is normal for an idler pulley clutch. You could probably stop them from turning by putting the tines into the earth, but I sure wouldn't want to try and stop them with my hands. I pulled on the hydraulic lever to raise the tiller off the blocks I had it positioned on and engaged the idler. It's alive!

The tiller is really quiet. Just the whirring of the chains and tines. Sounds really good. The rear belt was flapping as it's a bit longer than the front one. Both belts are in pretty sorry shape.  I have to get some new belts ordered, but hopefully the old ones will last long enough to try the tiller out tomorrow. I'd like to know that I assembled everything correctly and that I don't need to strip it down to fix some problem or another.

October 13, 2012
I decided not to try and run the tiller using the link rod that I use for my sleeve hitch. If I used it, there is no way to lock the tiller all of the way up in the transport position.  The cylinder that controls the rock shaft does leak down a bit and I didn't want to chance the tiller hitting the ground as I drove out to the area I am going to till. My choices now were to either notch the 140 link rod or fabricate a new one. I guess I could also wait for a 300 series link rod to show up somewhere, but the chances that I'd find one before the weather gets too cold are probably slim.

I decided that I would cut a notch out of the straight 140 link arm so that would clear the transmission mount. This would reduce the amount of metal in the rod by half, but seeing as there are two 3/16" pieces of steel strap welded together, it should be strong enough to test the tiller. Once I test the tiller operation, I will weld in a section to strengthen the rod and still clear the transmission mount.

With the modified link rod in place, I could now use the depth adjusting cam.  I raised the tiller and locked the cam into the transport position and rode out to the new garden area. I readjusted the cam and set the depth for about 2 inches. I fired the 317 back up and engaged the tiller. As I lowered the tiller into the soil, it pushed the tractor forward a couple inches while chewing through the short grass, weeds and soil. I slowly drove forward about 20 yards and stopped to examine my work. The ground was nicely shredded with the bits of sod and roots sitting on top of the soil. I didn't hear any bad noises coming from the tiller and all of the tines were where they had been welded.  So far, so good.  Another couple passes and I checked the tiller over again. Still good. I ended up tilling a section about 70 feet by 25 feet before one of the belts flipped off. I was expecting this but at least I was able to test out the tiller and it passed with flying colors.

Back in the shop and the tiller has broken some ground - as well as shredded some belts.	Long shadows show it's late in the day, but we now have a part of the new garden tilled.

In the tiller manual, I had read that when using both extensions one should not try to till full depth in one pass. As I learn what the tiller is capable of, I will error on the side of caution and try not to stress it too much.  I am trying to till fairly hard packed clay sod that has not been broken in at least 20 years (and maybe never) so being cautious is probably the best way to approach this job.

As I said, I need to order a pair of belts for the tiller.  Depending on the PTO style, Deere lists two: AM32172 and AM32636.  From what I can tell, the former is for the earliest solid-shaft PTOs and the latter for the rest. The AM32172 has a length of 35.4 and the AM32636 has a length of 35.76. In the belt cross-references I looked at, they both supersede to a 4L-350 belt.  Seeing as how my PTO is shop made, I'm not sure that three and a half tenths of an inch are going to matter. In the past, I have purchased all of my Deere belts from Deere. However, I can get two 4L-350 belts for $11 delivered in a day from McMaster.com and Deere wants $40. I'm all for supporting my local Deere dealer, but not this time.  We'll see how it goes.

I think that the PTO pulley looks pretty good peeking out from the fender pan.	New attachment stud bushing on the bolt and the old ones above it.

October 18, 2012
The belts arrived yesterday. As you'd deduce from the number, they were 35". So they're between a half or three quarters of an inch shorter than either of the Deere belts. They were a bit tough to get over the pulley and I ended up using the old mechanic's trick of using the starter to bump them on. However before I could put them on I had to deal with something I had forgotten to do.

The attachment studs that hold the bottom tiller mount to the bracket were too short. The existing studs had a 0.320" wide bushing. This fit the sleeve hitch perfectly, but the tiller mount was a little more than 1/2" wide. To test the tiller, I had just loosened the studs, but there's no way I would run it like that for more than a test.  The stud bushings are .75" OD and .5" ID.  I used some 1" drill rod to make the new bushings on the lathe.  Outside diameter was 0.75" with an 0.5" hole and 0.550" in length. I re-used the original bolts. I used a zip tie to attach the old bushings to the sleeve hitch so I wouldn't lose them.

The new belts are in a lot better shape than the old ones. You can see the modified position of the coupler on the spreader bar.	It was a bit of a fight to get the new belts over the pulley but they work great. The tension is set and the two belts seem to be exactly the same size.

In my test I had also found that the coupling nut I used to fix the spreader bar was in exactly the wrong spot. It rubbed against the depth adjuster cam and made setting the depth tough. I cut the spreader bar a bit shorter and re-threaded it so that the coupler was moved to the left a couple inches.  By this time it was getting late. No trying out the new belts tonight.

Late this afternoon I finally got a chance to run the tiller with everything fixed.  I'm impressed with how well it works.  I set the depth to about 4 inches and tilled the area I had tilled the other day. I then went back over it at full depth. It did a real nice job of cutting up the sod and mixing it in with the soil.  Tomorrow I will work some compost into the soil and let it be until spring.

The depth adjustment cam now has plenty of clearance from the spreader bar.	Blurry picture, but the tiller has done a great job of breaking up the sod and mixing it into the soil.

This tiller has not been the quickest or easiest repair/restoration job I have done, but seeing the tiller produce fluffy soil from the hard packed sod has made the journey well worth the time and money spent. As I hope will always be the case, I learned a few new things and got to use some tools. I can't ask for much more than that.

If you have questions or comments, drop me an email.

© Fager 10-07-12