Installing the Y axis ballscrews.

made the Y axis bracket on the C-Beam, later on turned out i didn’t have any M6 80mm bolts so we hastily cut it out of 1/8" acrylic. I also forgot to take off the "leave stock" option, but otherwise the holes it cut were the right size and decently round, so that’s a start. I’m going to recut it in steel today.

initially with the pulley blank from amazon, we lathed out the hole to fit on the ballscrew, and added a hole for a set screw which fits on the flat.

mmca then lapped and scraped the gibs, i didn’t get a lot of info on this one, but there are lots of videos elsewhere and he is skilled enough to do it by sight, basically hi-spot dykem and a granite plate, some laps and fine wet and dry paper.

test fit, so far so good. this is the large slot we routed out manually with the C-Beam and the dewalt 611.

after adding the x table the double ballnut was scraping the underside of the table.. sigh yet more grinding of the cast iron, not fun, i wore out the dremel battery and me, my fingers are still vibrating a day later.

fixed that mounted the table and X and Y motors onto the pulleys.

you can see our shame in the yellow acrylic, also note the direction change of the pulley, we had some discussion about how to do it, eventually went with putting the set screw into the teeth and countersink/deburr and it lines up nicely. we talked about removing some so it’d go over the threads, and even putting a thread to match the thread of the ball screw ends.

set the software up to do 5mm per turn and 2:1 ratio on the pulleys 10/20T.

after some testing we say about 0.003" backlash, which was disappointing, so some fiddling around testing and retesting and discovering flashcut can’t do enough precision in its backlash compensation we moved on to look at why.

The double nuts on the ball screws aren’t preloaded, which seems odd so basically what you have on the ballscrew with a double nut is two sets of balls in opposition to each other, so they’re pressing up against the inners of the ballscrews themselves, that way when the direction changes there is no slack ours did have some movement so we fixed that with a shim

later on we’ll try a belleville spacer etc.

after this , and some accidental switching of things, numbers weren’t adding up, so after some investigation on why it’d step one but not the other, discovered the pulley setscrew on the motor was loose ( again) this has been common on the flashcut for me, so they need some threadlock.

once that was all sorted out the backlash was down to sub 1000th’s/in ( sorry about the mixed SAE/metric)

got a move rate of about 140in/min, which is 100 more than the lead screws..

again the plan is to strip it after test fit and then put it back together with more pictures, we’re in the test and see phase of things. for the boxes of screws i’ve picked up recently, the fact we’re always missing some size is maddening, especially since its nearly all metric and local stores pretty much carry SAE only with a smattering of metric, i’d say amazon to the rescue but they’re even on two weeks delivery for some of them.

next is to install the X axis ballscrew and we should be able to reassemble it for test runs before teardown again, and we’ll probably tear it down again after that because we can remake the various mounts and so on.

Advertisements