Gearbox

A few observations on how, and how not to make a worm gear drive for a telescope. Since the major item is making the box, with the application elsewhere, this is in the metalworking section.

 
Nothing ventured, nothing gained, so I decided to take the gear hobbing setup one step higher. I decided to make a gearbox. Now I had a piece of 2 inch wide 1/4 thick aluminum bar, and cut that in half. I shouldn’t have done that, in a sense, because it led me to a number of interesting mistakes, and a few other lessons learned.

 
The first picture was as originally designed, the second picture is the one that works.

 
Firstly, a picture of the hobbing setup done as a test. This does work, and did produce a gear. So I took another chunk of aluminum, and used an existing 1/4 inch hole, and made a gear hobbing setup that was a bit more elegant, well, perhaps I thought so. I’m going to do something about that a bit later, say, put in some ball bearings. I’ll need them.

 
Now the second picture is of a new gear blank (the first one was just a section of bar stock. This new gear has a hub turned down for a setscrew, which is a bit easier than a keyway. The picture shows the gashes made by the 1/4 –20 tap as it cuts the first threads in the gear blank. I wanted this to work, because cutting the gear blank wasn’t really easy. I was trying to be cheap with the metal and used too little. I should have used a longer chunk.
Here’s the gear after one complete revolution. I hadn’t measured, and just trusted that the teeth would come out right. Other than a small blurble, they actually did. That was just lucky.

 
Now other than not measuring, I hadn’t made any serious mistakes yet, at least, none that I know of now. This shows the gear a bit further down the road. I use tapfree fluid to smooth the cutting. It really doesn’t dissolve the blue layout dye.

 

A different view of the setup. The large chunk of aluminum bar on the left is a collar to keep the shaft in place. Not very elegant, but it did work. I will replace this with a more elegant setup, and use some ball bearings to let the gear turn well. I’ll also widen the setup to accommodate larger gears. I may turn some that are up to 3 inches in diameter.

The last picture shows the gear as produced, which isn’t too bad. I kept this going at low speed, and that kept the positions relatively constant, not too much knock and stress, although cuts should be light, only a few thousanths or so. Fortunately, the mill does all the work, so it’s only time.

 
Now here’s the two plates of aluminum. They’re bolted together so the holes match, and one hole was already there, in one plate. So I ignored it (not a mistake). You can’t see the problem yet.

 
The gear is positioned to allow enough spacing for the worm. If things look a bit crowded, they are. However, it all fits so far.

 
Here’s how the worm is planned to fit. Notice that it nicely clears the bolt holes that tie the plates together. I have another idea in mind for the plates, though. (and that idea had its own problems, see later). There’s an equally small amount of clearance on each side of the worm gear.

 
What I did here was to space the two sides apart, and clamp the motor mount plate on top. This is a commercial clamping setup, and this is the reason we buy this kind of thing. It was extremely easy, and held well enough to drill the matching holes.

 
Tapping the holes. It’s easiest to tap straight through. I’ll drill out the top holes and countersink them a bit later, although I could have done it in one series of operations without moving the plate. Notice that the motor plate sits on top of the sides, relatively speaking. You might start to see the problem. I didn’t.

 
The worm screw would go across the far side, down at the bottom. You can see the bottom support, which is cut out of a piece of bar. Now this I did reasonably well. Without moving the drill bit, having drilled the hole to match where the worm would go (I took a piece of rod, pointed the tip, put it in the mill chuck, and ran it down vertically. Where it stopped was where to drill for the worm gear. The gear is pushed to one side.

 
A picture of the motor plate with the pilot hole drilled, (more later) and the worm end support mounted.

 
What I did, without moving the first setup, was to put the motor mounting plate on. The drill was set up to drill that support, so the hole drilled for the motor should align with the one for the support. This turned out to be about a #10 drill, because of the way I turned the worm down.

 
Drilling the motor hole. It’ll have to be reamed out a bit, but this isn’t bad. Go buy the silver and deming drill sets, 1/2 inch shank and will drill up to an inch hole. Works in aluminum without a problem. Saved me lots of problems. Produces lots of swarf. See the problem yet? Neither did I.

 
Here’s the motor plate mounted, it sticks up a bit, but that’s as designed. You can see the worm gear and the worm support in the back.

 
I turned down the end of the brass rod to make a fit into the hole. I’d really like a ball bearing for the shaft, but that may not be. The shaft extends out the motor hole, waiting to be trimmed to length.

 
Here’s the end of the worm shaft turned down. It matches the size of the shaft on the stepper motor, so one drill through will get both of them. There’s a bit of a problem here, but not the main one.

 
Here’s the motor mounted. The problem still might not be obvious to first view, but it is there.

The motor has screws that run through almost all the way and are tapped in the near motor frame. They mount the motor with screws that come in from the bottom into threaded holes.

 
However, the motor is so wide that the mounting holes line up very well with the sides of the gearbox. In fact, the motor covers two of the attachment holes for the motor plate, and the sides cover all but two of the motor holes. I can’t win with this one. So what I think I will do is to redo all the metalwork, put in the bearings I want, and other good stuff. It’s a nice proof of concept, but it isn’t sturdy, not at all. Also look at the shaft coupler. It’s really close to the worm gear itself, so close that I had to turn it back to get it to fit. I think I’ll try to make this a bit longer so I have more room between the worm gear coupler and the gear itself. I also want those ball bearings, and I might just put one on the stepper motor worm, too.

 
So here are the solutions:

mount the motor plate between the sides, not on top of them. This way, all I have to worry about is the holes running into each other.

Make the sides longer. I’ll have to, anyway, because the motor plate takes up 1/4 inch more this way.

Ball bearings for the driven gear. The shaft will come out 1/2 inch on the non telescope side, and I’ll put some sort of indicator plate there for rotation. The driven shaft will come out about 2 inches or so. It will drive the telescope through a rubber wheel. I’ll need another ball bearing support at the end because all the weight will be on that one shaft. (or at least 1/3 of the telescope).

The interior plate will be thinner, and will probably have a small ball bearing on it. I’ll extend that about 1/2 an inch beyond the gearbox walls, because that will be an attachment point where I’ll put another indicator wheel. That’ll show that the stepper is being driven and is moving.

I’ll also allow some space for the angle bracket mounts. I’ll need two of them to attach to the telescope base.

Moving the sides a bit further out gives me more room for the setscrews on the shaft coupler.

I’ll probably think of more stuff later.
And here’s what I tried:

1) Mount the motor plate between the sides, not on top of them. This way, all I have to worry about is the holes running into each other.

This is a good idea, but it had a problem, too. That extra distance had to come from somewhere.

The gear was so close to the sides that they had to be milled down a bit for more room

2) Make the sides longer. I’ll have to, anyway, because the motor plate takes up 1/4 inch more this way.

Now that worked, but I only made the sides a bit wider, they needed to be more than 1/2 inch because I lost the full half inch in getting the motor plates mounted.

 

3) Ball bearings for the driven gear. The shaft will come out 1/2 inch on the non telescope side, and I’ll put some sort of indicator plate there for rotation. The driven shaft will come out about 2 inches or so. It will drive the telescope through a rubber wheel. I’ll need another ball bearing support at the end because all the weight will be on that one shaft. (or at least 1/3 of the telescope).

So far: this has worked.

 
4) The interior plate will be thinner, and will probably have a small ball bearing on it. I’ll extend that about 1/2 an inch beyond the gearbox walls, because that will be an attachment point where I’ll put another indicator wheel. That’ll show that the stepper is being driven and is moving.

So far, so good. I don’t have a drill the exact size I needed to make the ball bearing fit properly, but I got close.

5) I’ll also allow some space for the angle bracket mounts. I’ll need two of them to attach to the telescope base.

Haven’t put them on, yet, but there’s holes already drilled for this.

6) Moving the sides a bit further out gives me more room for the setscrews on the shaft coupler.

And that was needed

7) I’ll probably think of more stuff later.

Still thinking about that one.

 
Now as I originally redesigned this, the long horizontal sides were vertical. Since they have 22 mm holes bored in them (limit of the 4 jaw chuck!) they were good to preserve. The problem was that the worm screw was so short that I couldn’t attach it to the stepper shaft. This was not a Good ThingÔ I had to turn the sides by 90 degrees, which happily made everything work.

 
You can see the driven gear bearings and the driven shaft. The shaft is 5/16 stainless, which is almost a proper fit for the 8 mm inside bearing races. Good enough, I suppose, for the experiment.

 
Inside view. The two shaft collars hold the shaft in place, they’re homemade. The gear is in place, and the bearing for the end shaft of the worm is more or less in place. I really want the boring head, I do. Still on backorder. I need a way to make precise holes in flat plate without the lathe.

 
Top view. Everything more or less lines up.

 
Now this is the shaft coupler from the first attempt. Notice that the holes for the motor mount are a bit egg shaped. This allows me to adjust for “manufacturing tolerances” and adjust the motor position so the worm fits the gear a bit better. The mounting holes are counterbored a bit, because the screws are short.

 
Another view of the motor holes. I’ve got to figure out a better way to align motor holes and align large holes. This works, but I’m not happy with fit and finish, here.

 
Evaluation:

Well, there are bearings were there needs to be bearings, and that’s good. The fit and finish and spacing issues were interesting, and luck played a good part of some of it.

I think this design is better, and it allows more space for the worm screw, which is needed. I’m happier with ball bearing assemblies installed, although I’m going to put in a precision sleeve for the gear, since I messed it up a bit. I’m waiting for a precision ream to come in, just don’t have the right size. I’ll leave the top open, and the next part will be to put angle brackets on the sides to allow mounting to the base.

Electronically, I need a stepper motor tester, and that’ll be the next thing. That will be a simple electronics project, and I’ll just use the lathe and mill for construction as needed. No complicated mechanicals here. I’ll post details after it’s built, but it will probably use a one chip microprocessor to accommodate various types of stepper, various pulse rates, and other good things. I guess I’ll build it as a general purpose piece of test equipment, and we’ll see what happens.

So far, the mechanical assembly is ok, but I’m realizing that I rather hate building gearboxes, at least, the ones that go badly.

Still, though, managed to salvage stuff, so this is good.