Lathe Debut Imminent
I’m not at church today because the A/C blew out. I can’t wait until we have our own building.
The CNC mini-lathe is pretty much done. I have to get the electronics going, but apart from adjustments and nonessentials, the hardware is ready to go.
I will not criticize this design, because I am not competent to do so, but I have had some negative comments from experienced CNCers. They don’t like the direct couplings between the steppers and screws, and one said it was a bad idea to use a lead screw to move the saddle. I don’t know what else you would use, though.
Yesterday I cut the lead screw to size. It had to be 22″ long, with a 1.75″ portion turned down to 0.25″ to receive an adjustable collar. I got it to about 0.248″.
Turning a lead screw in a lathe is pretty nerve-wracking. If you simply stick it in the chuck, you risk ruining the threads. You can bore soft jaws out so they hold it without marring, or you can make a split bushing that goes over it. You chuck the bushing, and it tightens around the screw without harming it.
I made bushings, but in practice, they were no good. When I chucked the screw, it wobbled pretty badly. I considered boring my soft jaws, but I wasn’t sure I wanted to bore all the way through a brand new pair. I might need that metal for something else later.
This was not a precision job, and I had extra screw material, so I tried the gonzo approach. I wrapped the screw in several layers of foil and chucked it.
It’s hard to put an indicator on a coarse thread. It will bounce around. I decided to eyeball it. I put something straight next to it and turned it, and I saw no deflection. I knew that put me within a couple of thousandths, which is not far from the best the chuck can do, so I was happy.
I center-drilled the far end, put a live center in it, and turned 1.75″ down to size. No problems. If it’s not perfect, I can’t detect it without instruments, and that’s more than good enough.
I had a problem when I mounted it on the lathe. I have some new flexible “plum” couplings, and I put one on the z screw. These couplings have an aluminum spider fitting at each end, and in between, there’s a layer of urethane, like a skateboard wheel. They allow for a lot of error in the rotation of the things they connect, but they also pull apart if you pull the ends away from each other. I want my setup to be rigid longitudinally, so the screw isn’t slopping back and forth and creating incredible backlash. I had to go back and use my old rigid connector, which works fine.
Plum couplings are perfect for the x axis, except that the one I got was a tiny bit too long. There was a layer of urethane between the shafts I was connecting, and it was too thick. In order to use it, I would have to make a thicker x mount or alter the coupling. I’ve made the x mount twice already, so my answer was to mill the urethane out. It should not matter. It installed easily, and it looks like it will run fine.
The z screw system goes motor, coupling, screw end, adjustable collar, wave (spring) washer, thrust bearing, mount, thrust bearing, screw body. The collar appears to be there to hold the screw tightly against the mount, putting pressure on the washer and bearings. It has to be very tight. I had problems with it slipping. I had to tighten it to the point where I was afraid the set screw would strip. I don’t know if 0.248″ is too small for the collar or what. When I made the screw, I assumed a couple of thousandths wouldn’t matter. It’s adjustable, right? But it may be that I need to put some foil inside it to take up room. Or maybe I need to make a bigger, tougher collar. In the real world, there would be a taper, I guess, with the bearing pressed onto it so it could not move. But this is not the real world. This is home CNC.
I don’t have a project box, but I have a Bud Portacab on the way. This is a foot-tall lunchbox sort of thing made from aluminum. I want to put plugs on the back for 8 motors (the limit for two Kstep boards). I’m not sure what kind of connectors to use. They have to be 4-wire connectors that will handle up to 18 gauge wire. I will also need a plug for the lathe’s spindle sensor.
I’m getting a lot better at machining, because I have no choice. It’s easy to machine when the plans are in your head, and everything is approximate. When you’re working from drawings, you have to apply yourself.
I’m hoping to see movement from the lathe this week. I need to get it put together and figure Gcode out well enough to do a simple cut. After that, things should get a lot easier. It will just be a matter of study and practice.
Most CNCers don’t do lathes, and that’s understandable. A mill will do a lot more. But I don’t have a mill waiting to be hacked up, and the mini-lathe was sitting here doing nothing. It should be very useful, though, and it will give me some understanding of the most important tool technology of our age. Most people are getting left behind when it comes to CNC/3D printing, and ten years from now, they’ll be like the old geezers who hate cell phones. “Damn this thing. Hello? Hello? What? I have to open it? Hello? Hello? Which button?”
After this, I have to make the lathe ball-cutter for which I bought plans. After that, I might like to make a small surface grinder. I considered making a tool grinder, but they’re not really that useful. People tell me that even if you have a great commercial grinder, you will probably do bad work, because it takes skill. And cutters aren’t that expensive, really. A surface grinder would be nice. If you like making machining accessories, you will want truly flat surfaces once in a while, and hand scraping is a real chore.
Hope I have a victory Youtube to post soon.