[ngphil]

Right now I've got the X tight enough for my needs. The Y is giving me issues. I believe the root of the problem is the cast nuts on the ends of the leadscrews that connect to the stepper shafts. They seem to be a fairly soft material and I think part of the problem is the surface that bears on the washer on the table has worn down which causes backlash in the screw between the collar of the screw and the back face of that nut. The Y axis still seems to have play in the nut inside the table and I've tried adjusting the backlash setscrew but it doesn't seem to help.

I was running a part last night in aluminum w/ a 1/16th endmill running at 2 ipm and taking .010" passes to get through a .125" thick piece of aluminum. The part was 2.5d we'll call it. By the time it got to the final roughing pass, the part was off by almost .070" in profile. That's because the Y seems to be slipping by as much as a .006 or more. It's hard to explain in writing but I can see it happen. When the stepper turns, depending on the direction, it pulls the nut away from the retainer washer on the table or into it. There's play there and then in the table. So on small moves, the stepper turns maybe 1/4 of a turn but that doesn't actually move the table. So while it "thinks" it's moved .006" (or whatever), it hasn't. That .006 (or more) error built up over 12 or 13 passes is causing the Y axis cut to be "stepped" rather than a 90 degree face. The X axis on the part was fine.

To make matters worse, today when I was trying to get the leadscrew nut tighter against the table mounting ring I broke a drill bit off about 3/4 of the way through the nut and leadscrew shaft. Now the nut is permanently attached to the leadscrew shaft. The problem is, I had the table mount on the leadscrew when I was drilling it and I just happened to have it on the wrong way (nut head toward the table side). Apparently those mounting rings are hardened steel . I just ordered 10 each of carbide spade drills in the 2 sizes needed to re-drill the table mounting screw hole in the ring. If I can successfully drill a new hole in the table ring then I can get it back on the mill. At least there shouldn't be much backlash there now! LOL I don't know where to get those nuts that are on the leadscrew shaft ends so if I destroy one I'll have to buy a new cnc conversion kit. BTW, does anyone recognize those toothed nut couplers? If I had a cnc mill I could probably make new ones LOL. Speaking of which....

In the meantime, I bought a taig this morning. I figured it was worth a try. At least it's a brand new machine. I ordered it from deepgroove1. I'll keep the sherline and try to get it fixed up. Once I get the taig I can at least make some parts to possibly get the sherline working smoothly. Then I'll have 2 mills to use anyway. If I can get them both making parts then that's actually very helpful in the long term. It sucked running those long programs and just waiting for it to do it's thing. It seems like the taig has much faster rapids. The sherline for some reason (crappy steppers?) doesn't like doing rapids any faster than 12 or so. I was looking at those syil machines. If I can make a go of what I'm doing maybe that'll be my next investment.

[ngphil]

Here's an image of what I've been using to test cut. I designed this so that a 1/6" endmill will work. In this part, when I tried cutting it last night, the horizontal grooves were stepped. It's not that the parts need to be insanely accurate, it's just that the buildup of errors results in a really badly formed result. I realize I don't necessarily need to run this at .010 levels, but that's really part of the test. While this part is fairly straight forward, some of the other things I need to do aren't and will require fine passes. If it can't handle doing .010 depth passes in straight lines then the other stuff is completely out the window.

[hoxsiew]

If your errors are building up, then that would indicate that the stepper is skipping steps.

[binfordw]

I wonder- if 10+ years later, if he sorted things out....