Loading ...
Sorry, an error occurred while loading the content.

Re: Rotary axis finishing paths

Expand Messages
  • Denver Hull
    Hi Rab, Thanks for the response. I used blender to make the models. When you say it s often better to make the splines directly, are you talking about
    Message 1 of 5 , Nov 2, 2008
    • 0 Attachment
      Hi Rab,

      Thanks for the response. I used blender to make the models.

      When you say it's often better to make the splines directly, are you
      talking about incorporating them as part of the initial model, or are
      you referring to the technique of manually creating the toolpath
      splines as opposed to letting the script create them?

      If I use the manual method, it looks like it's necessary to make
      separate line segments for each curved and straight area of the model.
      That appears to be how it's done in the tutorial. Does that give
      better results because each short segment of the initial spline can
      follow the contour better than an evenly divided mesh over the entire
      surface?

      Thanks,

      Denver



      --- In CNC_Toolkit@yahoogroups.com, "rainnea" <yahoo@...> wrote:
      >
      > Hi Denver,
      >
      > Either method would work, or project a helix spline.
      >
      > How did you make the models?
      > Often it's better to make the splines directly.....
      >
      > Rab
      >
      >
      > --- In CNC_Toolkit@yahoogroups.com, "Denver Hull" <denverh@> wrote:
      > >
      > > Hello,
      > >
      > > I've been trying to apply what I've been reading in the User Guide to
      > > a couple of my own projects, concentrating mostly on the rotary axis
      > > section. My ShopBot doesn't have an indexer yet, but it's getting
      > > close, so I'm looking at ways to generate the necessary toolpaths.
      > >
      > > If I understand things correctly, there are two general methods
      > > available to do this. One is to use the toolkit to produce an array
      > > of parallel splines, which then get projected them onto the surface of
      > > your model. The other approach is to project one spline onto your
      > > model by hand, then array copy it around, using enough copies to
      > > provide the desired stepover. Then use the toolkit to produce the
      > > toolpath for the projected splines. So far I've experimented with the
      > > first method only, but the Guide makes it sound like the second is
      > > better. I'm not sure I understand why that would be.
      > >
      > > I've uploaded two images in the Photos section in an album named
      > > Denver. I would welcome any suggestions for the best way to create
      > > toolpaths for these. Currently I carve the molding on two sides. The
      > > pieces are about 1" in diameter and 24" long. I use a 1/8" ball nose,
      > > with a stepover of .015". That gives a reasonable finish for
      > > something that's going to be painted and installed two stories or more
      > > above the ground. I don't make a roughing pass for these. Instead, I
      > > edit the toolpath file and set the move speed low for the first
      > > longitudinal pass, then set it back to normal. I haven't actually
      > > made any of the finials yet, but I don't think carving two sides is
      > > going to work for those. Four might, but an indexer would be best.
      > >
      > >
      > > Thanks,
      > >
      > > Denver
      > >
      >
    • Siwel Leinad
      ... Hi there, The helix finishing splines in the toolkit guide were made using the following method. 1. A Single helix spline was created in gmax such that
      Message 2 of 5 , Nov 2, 2008
      • 0 Attachment
        --- In CNC_Toolkit@yahoogroups.com, "Denver Hull" <denverh@...> wrote:
        >
        > Hi Rab,
        >
        > Thanks for the response. I used blender to make the models.
        >
        > When you say it's often better to make the splines directly, are you
        > talking about incorporating them as part of the initial model, or are
        > you referring to the technique of manually creating the toolpath
        > splines as opposed to letting the script create them?
        >
        > If I use the manual method, it looks like it's necessary to make
        > separate line segments for each curved and straight area of the model.
        > That appears to be how it's done in the tutorial. Does that give
        > better results because each short segment of the initial spline can
        > follow the contour better than an evenly divided mesh over the entire
        > surface?
        >
        > Thanks,
        >
        > Denver
        >


        Hi there,

        The helix finishing splines in the toolkit guide were made using the
        following method.

        1. A Single helix spline was created in gmax such that the diameter
        of the helix exceeds the largest diameter found in the 3D model. The
        Helix is of a single rotation. The length of the helix is set so that
        the single rotation occurs over the length of area to be machined by a
        single pass of the tool.

        2. The single helix was then projected onto the surface of the model
        in the X Axis. This conforms the helix to the surface. Suitable
        subdivision settings are used to allow for curvature. Effectively this
        creates a spline representing a single pass of the tool.

        3. The projected spline was edited and duplicated with the array
        tool. The edit is to enable easy and fast connection of the
        duplicates avoiding excessive tool retracts in the post-processed path.

        The above approach was adopted for a number of reasons.

        It involves the projection of only a very simple spline.. This is
        considerably faster than projecting a more complex spline based on
        stepover value.

        The gmax array tool can copy splines faster than gmax can project
        them. Stepover can be calculated within the array tool. Adjusting the
        number of splines in the array over a 360 degree rotation provides
        information on the resulting stepover.

        Hope this helps

        Danny
      • Denver Hull
        ... Hi Danny, Yes, that makes sense, I ll try it. I m still working through the tutorial, but I looked ahead a bit and found where the single turn spiral was
        Message 3 of 5 , Nov 9, 2008
        • 0 Attachment
          --- In CNC_Toolkit@yahoogroups.com, "Siwel Leinad" <yohoodi@...> wrote:
          >
          > --- In CNC_Toolkit@yahoogroups.com, "Denver Hull" <denverh@> wrote:
          > >
          > > Hi Rab,
          > >
          > > Thanks for the response. I used blender to make the models.
          > >
          > > When you say it's often better to make the splines directly, are you
          > > talking about incorporating them as part of the initial model, or are
          > > you referring to the technique of manually creating the toolpath
          > > splines as opposed to letting the script create them?
          > >
          > > If I use the manual method, it looks like it's necessary to make
          > > separate line segments for each curved and straight area of the model.
          > > That appears to be how it's done in the tutorial. Does that give
          > > better results because each short segment of the initial spline can
          > > follow the contour better than an evenly divided mesh over the entire
          > > surface?
          > >
          > > Thanks,
          > >
          > > Denver
          > >
          >
          >
          > Hi there,
          >
          > The helix finishing splines in the toolkit guide were made using the
          > following method.
          >
          > 1. A Single helix spline was created in gmax such that the diameter
          > of the helix exceeds the largest diameter found in the 3D model. The
          > Helix is of a single rotation. The length of the helix is set so that
          > the single rotation occurs over the length of area to be machined by a
          > single pass of the tool.
          >
          > 2. The single helix was then projected onto the surface of the model
          > in the X Axis. This conforms the helix to the surface. Suitable
          > subdivision settings are used to allow for curvature. Effectively this
          > creates a spline representing a single pass of the tool.
          >
          > 3. The projected spline was edited and duplicated with the array
          > tool. The edit is to enable easy and fast connection of the
          > duplicates avoiding excessive tool retracts in the post-processed path.
          >
          > The above approach was adopted for a number of reasons.
          >
          > It involves the projection of only a very simple spline.. This is
          > considerably faster than projecting a more complex spline based on
          > stepover value.
          >
          > The gmax array tool can copy splines faster than gmax can project
          > them. Stepover can be calculated within the array tool. Adjusting the
          > number of splines in the array over a 360 degree rotation provides
          > information on the resulting stepover.
          >
          > Hope this helps
          >
          > Danny
          >

          Hi Danny,

          Yes, that makes sense, I'll try it. I'm still working through the
          tutorial, but I looked ahead a bit and found where the single turn
          spiral was used.

          Let me see if I understand the rotational stepover calculations:
          Assume 1" maximum diameter
          Circumference = 1" * pi = 3.142"
          Desired linear stepover = 0.020"
          Number of steps per revolution = 3.142 / 0.020 = 157
          Number of degrees per step = 360 / 157 = 2.293 degrees
          Does that look right?

          Also, I've added another image in the "Denver" album, titled "rail &
          cap". This is a handrail and newel post cap, to be done as a single
          item. This one could be done either two sided (left & right, not top
          & bottom) or with indexer. However, the cap has a 3/4" hole in the
          bottom. I'm not sure how to handle that hole. If the top of the cap
          was flat, it could be done afterwards. If done with the indexer, it
          should be possible to define the location and depth of the hole and do
          that as a separate step. But I'm not sure how to get that done. Any
          thoughts?

          Thanks,

          Denver
        Your message has been successfully submitted and would be delivered to recipients shortly.