Building Jerry's Donkey

[Captain Jerry]

Just when I thought I knew How this winch worked, I found out something new. Here is a picture of the part where the rotation of the clutch lever applies axial thrust to a pin in the end of the shaft. I have looked at this picture for years and got it wrong. It appears that the part marked here:

Is part of the bearing support stand and that the shaft rotates inside of it. I was wrong. In the next view from a different angle and under different light, it is clear that it is not part of the bearing stand, you can see the fine line of separation.

It is fixed to the shaft and rotates against the bearing stand or against a bushing. This is critical. When the clutch pin is pushed in, it pushes the clutch shoes out against the clutch drum but some of that force is transferred to the gear face by the sliders, and the shaft wants to move in that direction as well. I knew that there had to be a thrust collar somewhere but I could not see it. So I had done it wrong. I had used flanged oillite bushings at both ends of the drum. They let the drum turn on the shaft, and they transferred the axial thrust to a flanged bushing in the stand at the far end of the drum. It worked, as I have demonstrated, but it was fiddly and difficult to adjust. It also added a lot of friction to the whole thing. It added enough drag that I felt the need to do some testing to be sure that the engine will have enough power to drive the winch at 200/300 RPM.

I had previously used my Unimat in a to drive the winch in trials, so yesterday, I rigged up a more solid test stand with the unimat drive.

Lowest speed with belt drive.

And reduced to about 200 RPM with a router control box. This box does not give good torque at very low speeds. It is easy to stall the chuck between your fingers. Not very precise but it seems to be in the power range of a twin cylinder air engine.

It took a little bit of fiddling to get the spindle and the crankshaft aligned and a little bit more time to get the clutch adjusted sot that it would engage and disengage smoothly. Playing with this thing is so much fun that I spent some time just playing. I noticed that when the clutch was applied, there was some flexing in the base and that the bearing stands flexed outwards as a result of the thrust being transferred through the winch drum. I was able to lift and lower a 5 pound weight with moderate pressure on the clutch lever but the Unimat motor wasn't happy. It would slow from 200 RPM to less that 100 RPM and get real jerky and stall.

I also noticed an irritating noise from the gears. They have never been run together for more than a few minutes and they had been in a box under the bench for over a year so they weren't bright and shiny. I didn't like the noise and I didn't like the flex so I took it completely apart and put just the pinion shaft and the winch shaft with gear back on the frame. I thought I would let the gears run in with a touch of Brasso and 30weight oil while I scoped out the clutch works.

In just a few minutes, the gears were running very smoothly so I let them run a few minutes more while applying some load with my finger on the big gear. I took them out of the frame and washed them with solvent and reassembled the whole thing for some more observation (playing). Gear noise all gone, smooth as silk. Flex and friction drag still there. I was starting to think that I needed ball bearing thrust washers at each end of the drum but that would be a real lubrication problem.

The quitting whistle blew so I cleaned up and headed to the house feeling a little a let down. After dinner, I was going through my "donkey" files, pictures and drawings, getting better organized to resume this project when I finally noticed the small separation line in the second photo above. Ureeka! That is the answer. That IS a thrust collar. All the axial thrust is absorbed right there. No transfer to the winch drum or to the bearing stand at the far end. I will still leave the flanged bushings at both ends of the drum to keep it in position but the contact between the flange bushing faces will be completely unloaded, even under maximum clutch pressure.

This morning, I turned up a brass thrust collar soldered it to the end of the shaft. I had to adjust some lengths and clearances but I got it all back together. WOW! What a difference. No flex, and without the flex, the additional friction caused by bearing misalignment as the stands flexed out of parallel was gone. The electric motor slowed only slightly as the clutch was applied, even when I increased the weight to 3 pounds. It worked so well that I played with it for another hour, up, down, float, feather, hold, faster, slower, heavier, lighter and I got the feel of the machine and I was not satisfied with the action.

There is no over-center or knuckle-in action. When the lever is at the full end of travel, the clutch linkage comes almost straight but due to the geometry, it will not knuckle in. It has to be held. If you release pressure, it doesn't immediately drop the load, but it begins to slip and soon disengages completely. There is no quadrant or notches to hold the lever either. On the real one there is a stop at the disengaged end of travel but there is nothing at the full on end of travel.

So I start thinking. I redesigned this part of the machine because I don't think that I can cut internal square threads at this scale (or any other scale). I used a slot, milled at an angle across the actuator instead of a full screw thread, creating a straight ramp instead of a helix. I could file of the end of the slot so there is a straight step at the top of the ramp. So I did.

Now when the pin that acts as the internal thread reaches that flat spot at the top of the ramp, it notches in and will stay there until the lever is pulled back. The action is smooth and positive and subtle. No hint of any binding. I did have to readjust the clutch because of the shortening of the inclined part of the ramp. The adjustment is a real pain because it involves taking the drum out of the stands and making small adjustments to the position of the gear on the shaft. I bet they had a better way.

Of course they did. Thats is what the threaded hickey is sticking out of the end of the actuator, a clutch adjuster!

And look at the way that clutch lever is clamped around the actuator. It almost looks like a repair by the local blacksmith but it is actually shown and described that way in the AmHoist Brochure. I guess I will have to try to make one. So I did.

I didn't get any pictures of the process because the batteries in my camera had died so I took this after I got home tonight. It's not exactly right, The clamp bolt needs cutting and the shaft should have been tapered and a little longer and a handle should have been turned at the end before it was bent but those are familiar processes, I needed to see what it would take to fabricate the flat, wrapped clamp. It was surprisingly easy. It took about an hour, but I'm slow when feeling my way in the dark. I'm going to do another one tomorrow. Better I hope.

I guess I'm officially back into this. After all, here is the first new part. Stay with me if you don't mind reading incredibly long posts. They won't all be this long but I feel the need for a full explanation when dealing with things that are out of the ordinary. I hope this explanation was understandable, I don't think pictures would make it any clearer, so I'll try to post a sectioned drawing of the winch shaft with thrust collar soon.

Thanks for watching, comments are appreciated or at least tolerated.

Jerry
.

 

[tel]

Like the frog on the windscreen - I'm still here!

 

[steamer]

......ribbit....ribbit


Dave

 

[metalmad]

Me too :big:
Pete

 

[Brian Rupnow]

Jerry---Welcome back onboard!!!---Brian

 

[dalem9]

I like it all , keep up the good work . Dale

 

[Captain Jerry]

A small crowd is assembling. I hope I don't let you down.

Here is the complete hoist drum assembly in cross section. I think it is better than a picture because you can see the works. The thrust collar (yellow) is soldered to the shaft, flush with the end. It bears against an oilite bushing (red). The bushing is held in the bearing support and transfers all thrust to the support. Pressure on the end of the pin (red) is transferred to the cross pin (green) which is fixed in the clutch spider (blue) which acts on the links (pink) which expand the friction shoes. There are two oillite bushings internal to the winch drum that let it rotate freely on the shaft and one on the far end of the shaft that is supported by the stand at that end.

This is probably the final design for this assembly.

Jerry

 

[Captain Jerry]

Here is the second try at the clutch lever. The cold formed. You gently urge the material from one shape into another shape while it is cold. That means you can whack it with a hammer as many times as you want but no heat is allowed. The no heat thing is kind of misleading. If you take a piece of steel, put it on an anvil, whack it with a hammer a few times and hold it between your fingers and you will get the point.

To make this work, you must chose your material carefully. Not to hard, not to soft. No music wire, no solder. I have found that SOCof works just right. My supplier stocks in rod in various sizes, from about 1/16" dia up to about 5/16" diameter, with several finishes. I prefer the "bright" or uncoated finish. Avoid the galvanized or the really weird glue coated. Strangely it is only available in short lengths, with a flat disk already cold formed on one end and a crude point on the other. The even thickness and regular shape of the cold formed disk is an indication of the material's cold forming acceptance.

For the clutch lever, I chose a size of a little less than 3/16" diameter about 3.5" long, Identified as size 16d. The first operation is to remove the disk and the point so into the 3 jaw chuck on the lathe to turn them off. Here is the first step:

It is then reversed in the chuck and the point is faced off and the end is center drilled for the live center. with about 1/2" is gripped in the chuck and the other end supported by the live center, a handle is formed at the tailstock end and the rest of the shaft is tapered using a 2.5° offset of the topslide. Here is what it looks like:

The finish is not very good but it will be improved by centerless grinding. That's what it is called when you roll it between thumb and finger while holding against the bench grinder wheel. Sorry, I didn't get a picture of that operation. Its hard to get a good finish while looking through the view finder.

Next is a view of the part positioned over the cold forming die. From here on, the process becomes a little precise. Whack it with a hammer over and over while attempting make the end of the rod flat evenly. There is a little technique involved in hammer whacking. If the part is starting to bend away from you, the near side is getting thinner. Raise your wrist a little. If it is bending toward you, the far side is getting thinner. Lower your wrist a little. Stop when the flat part is about 1/16" thick.

This is what it should look like:

Now the edges of the flat part should be filed to bring them straight, parallel and about 3/16"wide. It is a little easier if the coated with mark-up dye and scribed. Here is the part held in the bench vise with aluminum soft jaws:

and here is what it looks like after filing:

The next step is to bend the flat part into an eye with a 3/16" ID. More cold forming, a custom die is called for. The die is fabricated from a bar of CRS 3/16" thick and 1" wide. A 1/4" dia hole is drilled near one end of the bar leaving about 3/16" material between the edge of the hole and the end of the bar. Then the end of the bar is rounded over with a file. The profile of the bar end should be a full semi-circle. It looks like this standing up it the vise. The length of bar standing above the vise should be about the length of the clutch lever.

This is the part being formed over the die:

What's the hole for? This:

The final closing of the eye is done between the soft jaws in the vise. A 3/16" rod is inserted in the nearly finished eye and the eye is closed around the rod between the jaws. It is hard to get a good picture of this but I tried:

After the eye is formed, the handle is bent 90° to the shaft.

A little filing, a little buffing, clamp bolt (#2-56) and the result looks like this:

Much better than yesterday's effort. I'm satisfied.

Here is a video of it installed on the winch and demonstrating the knuckle in action that I worked out yesterday. When the clutch is engaged, it stays engaged by itself until disengaged.



And thus ends today's fun. 21520

Jerry

PS: The steel classification of SOCof indicates Somebody's Old Chevy or Ford

 

[Maryak]

To make this work, you must chose your material carefully. Not to hard, not to soft. No music wire, no solder. I have found that SOCof works just right. My supplier stocks in rod in various sizes, from about 1/16" dia up to about 5/16" diameter, with several finishes. I prefer the "bright" or uncoated finish. Avoid the galvanized or the really weird glue coated. Strangely it is only available in short lengths, with a flat disk already cold formed on one end and a crude point on the other. The even thickness and regular shape of the cold formed disk is an indication of the material's cold forming acceptance.


Jerry

Very Droll Rof} Rof} Thm:

Best Regards
Bob

 

[Brian Rupnow]

Nice handle Jerry!!!! I just read your recent post over a dozen times trying to figure out what was said-----I know its very plain to you. Basically it seems that all you have really changed is that you have installed an outboard collar on the end of the shaft, to bear against the outside head of the shaft support bushing which is "fixed" in the support housing. This takes up all of the axial thrust so that none is tranfered to the winch drum nor to the far end shaft bearing and support. Do I have that right?---Brian

 

[Captain Jerry]

Brian

That is exactly right. All I changed was adding the collar. A very small change that makes a very big difference in the way that it works. You have to understand that this is not my design. I am trying to build this to operate as closely as possible to the original American Hoist & Derrick design and that was one small detail that I did not get right. All I have to go by is a rusted hulk that sat submerged in salt water for about 50 years. There is also an original sales brochure that includes some drawings (engravings?) but these are not detailed technical drawings. There are two drawings that show the thrust collar. In one drawing, the artist got it right but in the other he got it wrong and his confusion added to my confusion. If I confused you with by description, well, there seems to be a lot of that going around.

Jerry

 

[Captain Jerry]

Oops. That's not all I changed. I modified the screw thread that drives the thrust pin so that there is a knuckle-in or over center action. I'll try to post a quick drawing to explain. It will take a few minutes to get it posted and I will modify this post with an attachment.

The drawing/model has been added. This is the part that is attached to the handle and is rotated against a pin which sits in the slot. The reaction against the pin drives this part against the clutch actuation pin. The feature that was added is the flat spot shown in red. It gives a spot at the end of the rotation, when the clutch is fully engaged, for the reaction pin to rest on. It prevents the clutch from disengaging when the handle is released. On a long haul, the operator could set the clutch and then roll a cigarette (a two handed job for most) while the log was dragged to the landing.

Is that clear? 21938

Jerry

 

[Captain Jerry]

Fire door is probably best done by squashing a suitable piece of tube and flycutting the end with a boring head to the boiler OD plus the thickness of the flange. Bend the flange to fit the boiler and then silversolder onto the curved end of the tube. Finnaly cut off excess tube, file away the flange plate inside the tube and an equal amount all round the outside then rivit to boiler.

Jason

You posted the above reply on Brians thread but I wanted to reply here so as not to confuse the two projects. The fire door has been staring back at me for years and you made it sound so easy I had to give it a try today:

That was easy! Thanks for the tip. There is more to do but the biggest hurdle has been passed. I wish I had asked for help before.

Jerry

 

[Brian Rupnow]

Looks good Jerry. ---Never be shy about asking for help on this forum. There are a lot of remarkably talented people who "see" things from a slightly different angle, making you wonder "Now damn---That was so simple!!! Why didn't I think of that???"

 

[Captain Jerry]

Well, here is the progress that has been made in the last few days. I'm not going into very much detail because much of it has been show before.

The hoist bearing stands have been re-profiled to reduce their bulk and give better proportions. Just spent some time on the mill.

Then I looked at the cylinders. I either had a poor plan or no plan at all for mounting them on the deck frame. Reluctantly, I decided they need to be made over. I like the look of brass but I plan to paint this project and except for a few parts, I need to starting to see this in BLACK. So the new cylinders are made of aluminum with an integral mounting flange, just like the original AmHoist design.

Using methods that I developed for the cylinders on the vertical and horizontal over crank engines, It starts with two blocks of aluminum on the left, with some profiling on the lathe and the mill, to the finished cylinders on the right. Yes, that is an aluminum cylinder, with brass heads and brass clad with black showing through. I think I like the contrast even better that I did the previous all brass.

Now that I can mount the cylinders, I can turn my attention to the Cross head guides. AmHoist used a single flat bar under the cross head. The mounting and aligning of this bar gave me fits before but now it seems very straight forward. The bar that I am using is cast iron salvaged from a broken drill press vise. It is suspended from a brace on the frame under the back leg of the bearing stand and at the other end of the bar it is suspended from a bracket on the cylinder head.

Here's theirs:

Here's mine:

Work continues. 22281

Jerry

 

[Jasonb]

See it wasn't that hard You are making good progress and have a good scale loking model there.

J

 

[Jared]

All right! Good to see this picking up again. I usually do things steady by jerks myself. Get burned out, do something else, half a year or two or five go by and I pick it up again. Looking forward to more.

 

[Captain Jerry]

Jared

Thanks for watching.

I have been having difficulty staying focused and here is another example. A few days ago, I got distracted from my plans by a couple of post that showed me how to go about fabricating the fire door. I had not even been thinking about it until Brian mentioned it and Jason showed the way. I didn't quite finish it but I got it in good enough shape That I proved to myself that I could do it.

This morning, my plan was to finish up the steam chest and covers so of course I took one look at this:

and said "that's not a fire door, that's a fire hole. Where is the door?" So forget the steam chest and covers and make a fire door. I still had about 1/2" of the squashed pipe that was used for the first part, so it was no big deal to make sure that I had a good flat end and then soldered it to a brass plate. Then sawed of the end with the plate plus 1/16" of the pipe. (Slitting saw in mill). Then a little cleaning, filing and finishing and here it is:

Now it's a fire door. Do you think it needs hinges? 22586

Jerry

 

[Jasonb]

Now it's a fire door. Do you think it needs hinges?

Oh Yes.

And a nice ring of rivits around the flange, then while you are at it a handle wouldn't look out of place

J

 

[metalmad]

Nice one Jerry :bow:
Pete