CNC controller

Discussion in 'CNC machining projects' started by ddmckee54, Sep 12, 2019.

  1. DavidF

    DavidF Administrator Staff Member Banner Member

    Well hop to it!! ;):p:p
  2. Gippeto

    Gippeto Silver

    X2... :)

  3. ddmckee54

    ddmckee54 Silver


    I've been researching homebuilt CNC routers for some time now and I've seen quite a number that were screw driven even if the screw was just a threaded rod, and some of them had no bearings on the motor end. I guess they figure the motor bearings can handle the load?

    I had planned on at least one step above that, a flanged bearing on both ends of the lead screw with ether jam nuts or locking collars to lock the lead screw into the bearing. As far as couplings are concerned, I've seen everything from the fancy-schmancy one piece spiral machined zero backlash couplings that will tolerate a few degrees misalignment, to one piece couplings that were fabri-cobbled out of an old bolt with a hand drill. It seems to me that as long as you have the centerline of the motor shaft and the centerline of the lead screw aligned, then there's nothing wrong with a solid coupling.

    As far as hijacking the thread is concerned, if you tell me about the controller and why you picked it, then you can hijack the thread all you want.


    P.S. - I just found this video on reducing the whip on 8mm diameter lead screws that are 1500mm in length.
    Last edited: May 4, 2020
  4. DavidF

    DavidF Administrator Staff Member Banner Member

    Dont think the bearings in a stepper motor would handle axial load very well and would probably have a ton of lash...
  5. Gippeto

    Gippeto Silver

    With the 3018, the stepper takes all the loads...solid couplers and no nuts on the bearing ends. Not saying it's "right", just that it "is". Think it's a poor setup, but it's a $200 machine.

    Re whip...came across this today;
  6. Al2O3

    Al2O3 Administrator Staff Member

    Maybe so but I'd never do it that way.....just me.

    The actuator I built on my pin router is acme lead screw driven. It has two preloaded radial contact bearings captured with jam nuts. The whole 3 1/2HP router carriage hangs on that thrust block and the end of the lead screw is unsupported beyond the antibacklash nut. It has a coupling to a gear motor.

    1003-1.jpg 1000-1.jpg 1001-1.jpg

    I wouldn't select an 8mm lead screw to drive that inertial mass, at those speeds, for that unsupported length.

    Don, I wouldn't get very far into that conversation before I was telling you more than I know.

    I didn't pick the controller because it came with a bundled system I bought and assembled. In addition to foam patterns for castings and typical router tasks, I'm also an avid fabricator so it's a plasma router hybrid. Even though in the long run I'd probably have been more satisfied building my own and will ultimately further modify this one, I want to use the machine and have the resulting parts more than I want to build a CNC system, and I still need to become a proficient solid modeler for the parts I want to make. I don't just want to cut profiles and 2 1/2D parts. I need to cut complex blended surfaces.

    Assembled Table.JPG

  7. ddmckee54

    ddmckee54 Silver

    That's one of the reasons why I plan on never using them to support a lead-screw.

    Gippeto - I like that set up, it's a little beyond my capabilities right now. but it is very neat.

    The video that I have a link to in post #23 shows the results achieved by adding a little tension to an 8mm diameter lead-screw that's 1500mm long. For as simple as the fix was, the reduction in whip was impressive. I think this definitely puts lead screws back on the table for consideration. Now I've just got to find a pair of reasonably priced lead-screws that are at least 1500mm long.

    I've got my eye on some 3/8-12 Acme screws that are 72" long, and they don't cost an arm and a leg. They are only single start screws, I've read that multi-start screws help to reduce whipping. However at this point I don't want to invest more cash money in lead-screws than I've got in the rest of the machine.(Maybe two or three routers down the road I will, but not now.) I'd use a flange bearing on each end of the screw to support the screw. I could use some 3/8-12 Acme nuts to both lock the screw into the bearings, and put a little tension on the lead-screw. I know this is not the proper application for these bearings and the accuracy of the machine will suffer. I'm not looking for 0.01mm accuracy/repeatability, at this point I'd be happy to get within a couple of tenths of a millimeter. I could either belt drive, or direct drive the screws from the stepper motors.

    Last edited: May 5, 2020
  8. boroko

    boroko Copper

    Kelly, did you cast the parts for your pin router?
  9. Al2O3

    Al2O3 Administrator Staff Member

    Yes, 25 years ago. They were all made with wooden patterns and cast in green sand at my friends foundry, except the grey iron tables which I had a small iron foundry in the area do for me.

    I talk about the most recent additions here. Good grief, even that was 5 years ago.

    This was one of my more ambitious projects with the pin router.

    Best to take up any further discussion in one of those threads, since I hijacked this thread of Don's long ago.


Share This Page