Welcome to Our Community

Some features disabled for guests. Register Today.

      1. Build Progress:
        • Build in Progress...
      Rate This Build
      0/5,
      0 votes
      Hi,

      so I saw the post a few weeks ago about the CNC competition. I've been hesitant to enter due to time constraints. I felt to enter I should sit down and spend some time developing a cohesive layout of my design, how it was to build, changes, results, etc. However, if I keep waiting for enough time to do this, I think the competition will have come and gone. So I thought I would set this up with what I can think of now, and hopefully I can add/subtract in the future.

      I'm a little unclear on how this will work. I asked in the contest section if the CNC's need to use Openbuild parts. I did not get total clarity, but I see that someone else asked a similar question and the response was "Does not have to be OpenBuilds parts ", so I assume this is acceptable...?

      This is one of two CNC's I've designed and made. I started designing this about a year ago, maybe more. Where I am located, I don't have a lot of readily available access to metals parts, or metal part shaping tools. Nor am I skilled with metal working. So my approach here was to use what was at my avail and within my existing knowledge and skill set - plywood.

      Please keep in mind, I did not know about OpenBuild products at that time, and they were fairly new to Canada - not available where I am located without ordering in.

      --------------------------

      Basic materials:
      1. SFU1605 ballscrew + SBR16 linear rail package
      2. Two sheets of Hardwood based plywood
      3. Screws, nuts and bolts
      4. Epoxy
      5. GRBL Raspberry pi Hat CNC controller
      6. Stepper Motors
      7. Wire (18 AWG)
      8. Time

      ...

      9. Accurate Straight edges, squares, etc
      ----------------------

      When planning out this project, I planned everything to be based on straight cuts, then to screw/glue those pieces together accordingly.

      --------------------------------------

      The gantry is comprised of:
      -3x 7" wide plywood piece
      -2x 5" wide plywood pieces
      -2x 6" wide plywood pieces

      The 7" and 5" pieces form an I-beam shape for the gantry. The side supports are created by a 7" and 6" piece connected together at a 90* angle. Originally I had intended for the gantry to use one ballscrew, underneath - with a long 6" piece running between the two sides, connecting them together. However, I found this to be difficult to implement and of poor results, so I opted for two ballscrews, one for each side.

      ---------------------------------------
      The main base is comprised of:
      -2x 4" wide plywood pieces
      -2x 5" wide plywood pieces
      -1x 30.5"x45 15/16" plywood sheet
      -Additional Reinforcing brackets
      -Furniture Bolts


      The 4 and 5" pieces form a square outline. Once this outline was tuned, I 'locked it' in place using the furniture bolts. Then, bolted down the plywood sheet and got to work adding reinforcements to lock everything in place.

      -----------------------------------

      A big part of this project was adding the mechanical components. Half of the mechanical rails came bent - up to several mm. Putting the machine together meant juggling several variables:

      1. Flattening the X-Y-Z faces
      2. Making the Linear Rails Straight
      3. Making the Linear Rails Parrellel
      4. Making everything square

      I found that I could reasonable make things flat, straight and parrellel, but not square. This sounds odd, however, square is not actually very important. What I mean by square however is this:

      _______
      >>_______

      Normally you would want the two lines to be parallel, straight and 'square' such that one does not extend past the other. However, I accepted that in my alignment that they would not perfectly provide corner-to-corner squareness. The consequence in this decision was a minor loss of a few mm of rail movement in each axis. A reasonable price to pay.

      It still took me 20+ hours to obtain accuracy of 0.05mm or better per axis. In reality, the X axis was within 0.02 mm, except for one spot where the rail had a surface aberration. In that one location, the axis went out of straight and parallel by as much as 0.05mm, but this was unavoidable and unrepairable.

      The X axis took the most time for two reasons.
      1. Unlike the Y and Z axis, I could not simple 'flatten' the face for both linear rails to provide a good mounting substrate. Both axis needed to be independently square with the plywood base sheet and with eachother and parrellel and straight. This proved very difficult, and the final solution implementation took more than 20 hours to achieve, after at least three similar iterations that used methods that failed. Ultimately, the solution involved making a very large jig.

      2. The Base X axis, having two ballscrews, required additional considerations for each ballscrew.

      -------------------------------------------------------------

      After extensive tuning, once each axis was in the correct position, I permanently locked it down through epoxy.


      ---------------------------------------------------------

      Unfortunately, I didn't take a lot of photos as I built this. I was usually focused entirely on what I was doing, rather than documentation. Also, I notice the spoilboard makes the X axis baseplate look terrible, so maybe I will remove it for some better photos going forwards. The advantage to this design approach is that the build process is fairly self evident, I hope. If not, I will add more details.


      ---------------------------------------------------

      For now I'm going to mark this as "incomplete". However, I did do some tests with this build, using a marker. It did great. However, I must have bumped something because I see the DRV8825 drivers are now damaged. i plan on Fixing those before too long. When I do, I will do some engraving or routing tests and then mark this build as complete.


      -------------

      Test video from: Feb 1st, 2018



      Testing it out using the design from here
      1. Special Notes

        It occured to me that this design could be implemented almost entirely with Openbuilder Aluminum rail components. I would advise anyone considering this build to replace all plywood components with aluminum ones, it would enable faster construction and better rigidity!
      koumac1 and Scotty Orr like this.
  • Loading...
  • Build Details

    Build License:
    • CC - Attribution NonCommercial - No Derivs - CC BY NC ND

    Reason for this Build

    Because I wanted to? Gluten for punishment? Who knows.
  • Attached Files:

  • Parts list

    Qty Part Name Part Link Comments
    2 SFU1605 750mm Link
    1 SFU1605 300mm Link
    1 SFU1605 1150mm Link
    2 SBR16 - 1100mm Link
    2 SBR16 - 700mm Link
    2 SBR16 - 300mm Link
    12 SBR16UU 16mm Link
    1 1.75 Hp Wood Router Link
    6 mechanical end stop sensors Link
    4 100 ft 18 gauge wire Link
    4 DRV8825 Pololu Drivers Link
    1 Raspberry Pi Link
    1 Raspbery pi CNC Hat Link
    1 1 litre two part epoxy Link
    2 4x8 Sheets of Plywood Link
    2 18mm x 37mm Openable Plastic Cable Drag Chain 1M L Link
    0 Link