No announcement yet.

How does one mill a 60" radius without a really huge rotary table?

  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    While this method is fairly well known and works just fine for generating a SPHERE or a part thereof, it does not generate a cylindrical shape and can not be easily modified to do so. The part in the OP was described as being 1/2" thick and the same radius will be generated across that direction as along it's 15" length where it is actually wanted. This may or may not be important for the part in question, only the designer can tell us for sure.

    According to my calculation of the sagitta of a 1/2" long segment of a 60" radius, the 1/2" dimension will be about 0.0005" deeper at it's center on the concave side and about the same amount taller at it's center on the convex one. This may or may not be OK.

    As I said before, TOLERANCES are required to actually determine a method for making this part. And a proper drawing.

    Originally posted by Denny98501 View Post
    Long radii are actually a lot easier and a whole lot faster if you don't have CNC.
    Analog is the way to go for a wonderfully smooth curve instead of a thousand tiny little steps.
    You need a manual milling machine, a fly cutter and small rotary turntable.
    Below is a photo of a 200 inch radius of curvature polishing tool that we made.
    Below that is a reference showing how mirror makers produce long radius curves.
    After 15 minutes of setup, it took about 10 minutes to rotate the turntable and produce the curve.
    The turntable is tilted the amount of the Sag or calculated as shown below.
    If tilted the other direction, a concave curve will be made.
    If you only have a small fly cutter, the curve can be generated in two passes, but the angle will have to be recalculated.
    Hope this helps,

    Click image for larger version

Name:	17 inch flycut tool 1024.jpg
Views:	175
Size:	125.7 KB
ID:	1899452

    Click image for larger version

Name:	Spherical Generation 001.jpg
Views:	173
Size:	147.8 KB
ID:	1899451
    Paul A.
    SE Texas

    Make it fit.
    You can't win and there IS a penalty for trying!


    • #47
      Originally posted by 754 View Post
      it could easily be done by swinging a router or circular saw , on a pivot bracket, even made of wood.. without a milling machine even.
      Glad you said it. Not sure but I think it was more of a class on lets see all the ways to do something.

      Its a very simple router job. But that doesnt work in the Machinist mind LOL

      I have 1/2" solid carbide bits that would do it. Slow going? Sure. Turtle gets it though... JR

      My old yahoo group. Bridgeport Mill Group


      • #48
        Unfortunately the OP has not responded to the question of tolerances.


        • #49
          Originally posted by CalM View Post

          How would you produce the wooden pattern =/-.002?

          Same problem as making the part. Just producing the layout lines is half the battle.
          Use a router with an arm to make a slightly longer form. Then rough cut the SS as close to shape as possible, bolt 3 bearings to the table and take light cuts on the mill while keeping the form against the bearings. 2 bearings to ride on the outside curve of the form, 1 to ride the inside. MDF may be sturdy enough, but HDF would be stronger. Either should be sealed.


          • #50
            Sorry about not getting back sooner... I couldn't answer some of the questions as I didn't know the tolerances. I had sort of assumed +/- 0.005 on the arcs since he had specified them to .0x So I sent an email asking for clarification.

            First of all, this thread has become totally theoretical. They found somebody with a CNC machine to do the job.

            Secondly, the part mounts on the bottom of a long swinging arm on the scope and is somewhat weight bearing. It rests against a series of bearings, and they wanted it to be +/- 0.003 (or better).

            I'd like to thank everybody for their input. I learned a bunch, and got some good ideas. My approach was going to be to make a jig that would have a series of 9 drilled and reamed holes using the DRO. Drilling and reaming 9 holes with +/- 0.001" accuracy using a DRO didn't seem terribly daunting (although I know that's near the limit of my ability). The jig would be registered onto three dowel pins on the table portion of the jig. The idea would be to move the jig carrying the steel along in small increments, and cut 2" long straight facets along my mill's X axis on the outer face (62" radius). The corner of a 2" facet will be 0.016" high compared to the desired circle. See the screen snip of a Fusion 360 design plan... blue = aluminum carrier plate on the jig, yellow = SS, and the dashed square is the "cutting line"...

            I hadn't decided if cutting down the 0.016" points to the desired line would be sander and file work, or if I'd move the jig so as to trim the tops off of those high points. That would give me a lot of little facets that would be 0.006" off of the desired circle. I was going to go file a bit of 1/2" SS and see how much bother it was going to be...

            However... it turned out that CNC won... (not that I lost anything ... it was one of those "six pack of beer" jobs for people I know).
            Attached Files


            • #51
              Well, at any rate this looks like a clever set up, demo just using a wooden board but I can imagine using something like that, perhaps from steel or aluminum and using clamps to attach the part to be milled, then you can cut one side of the radius, move the table over along the X-axis a specified amount and cut the other radius:

              I guess the tricky part is fine tuning the radius. I need to rewatch it later with sound.


              • #52
                OMG can't that be fit into about 45 seconds?


                • #53
                  I think there is some overlap between youtubers and people who keep on talking to the cashier in the supermarket line.