Announcement

Collapse
No announcement yet.

degree table

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • degree table

    There have been many ways suggested to divide a circle into various divisions. Here's another.

    Take a piece of all-thread, roll it into at least a circle and a half, then trim the first few inches off one end. What remains is something more than a full circle, with at least 360 degrees of it equally bent so the curve is the same anywhere along it. Ok, If you do the math, you can choose how many threads it takes to make the full loop and determine the diameter of the circle required to give you the number of threads you want. Each thread is one division. Cut the other end of the loop so you are one or more threads short of what you're designing for. A section of the waste curve can be used to align the ends of the loop to the point where there is no inconsistency in the thread spacing, and the desired number of division points is correct. It won't matter if there's a gap between the ends of the loop- the indexing 'pawl' will be a section of the waste curved all-thread that is at least several threads long, and will 'lock into' the threads anywhere along the loop, including across the gap.

    Now the loop will have some exact diameter. Math works great for determining this. Mill a circular slot in the bottom of your soon-to-be division table using some form of pivot point milling, and insert the loop into it. I'm skipping some of the details where you check, modify, check again, etc to make sure the loop is embedded in the slot with perfect spacing between the threads at the gap. When you know it's fitting right, lay some epoxy into the groove all around and press the loop into it. Lay the curved piece across the gap and clamp it. You'll be able to tell when it's right- it basically either fits or it doesn't. Clamp the rest of the loop until the epoxy sets.

    I've left out a lot of details, including the fact that I actually did this tonite- so if anybody howls at me for this hairbrained idea, I'll just say that a friend did it, and I thought he was crazy too.

    Soooo- as I type the epoxy is setting. I've wanted for a while now to make a round table that can index per degree, have a very low profile, and be usable flat on the mill table. This one is a bit under 7 inches across, made from 5/8 thick pvc, and uses 1/4-20 threaded rod for the loop. That means that for 360 divisions, this loop had to be close to 5.5 inches ID., actually 5.729 inches would be its pitch diameter. I selected the rod for the best looking threads, meaning no visible transition marks where the rolling machine might have done a shift of some kind. The loop does appear to have a very consistent thread spacing, something I was worried about.

    I'll probably make the 'pawl' adjustable for about a full degree, so I can get half degree settings, etc.

    One of my first uses will be to make a few angle accessories for the table saw. I sometimes need to make a cut at odd angles, and if it's 45 or 30 degrees I just use one of those plastic triangles to help me get it right on. The other day I needed a 20 degree angle, sometimes 22.5, and 43 degrees a few days ago. I'll be making up some custom triangles. This will work great for etching the degree markings on my pano mount, and the custom angle finder for at work. I can see it being useful for small scale gear teeth as well
    I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

  • #2
    Sounds good for a woodworking tool. How did you "roll" the circle
    without messing up the crests of the thread?
    If you used the short peice as a worm on the outside of the circle
    you could even sub-divide the one degree finer.
    [ a rotary table anyone? :-) ]
    ...lew...

    Comment


    • #3
      Darryl,

      You are giving me some ideas re a 4th axis for my mill. I can see having two separate devices that are specialized for horizontal and for vertical indexing instead of a "one size fits none" arrangement. A very low profile lazy susan type of table could be pretty useful if it were strong and stable enough to take a vise. I think I would go with a teflon sheet as the bearing and maybe belt drive it on the perimeter. Maybe use some sort electromagnetic clamp that can be actuated by the clamp code to hold it in place after indexing. Hmmmm.
      Free software for calculating bolt circles and similar: Click Here

      Comment


      • #4
        I'm going to have to use some clamping device of some kind also, since I'm not willing to rely on the indexing pawl to keep the table from rotating. In my case it's strictly mechanical and I just tighten it by hand as required. I like that idea of the cnc'd rotating table, the extra axis. One day I'll get closer to cnc myself.

        Part of my thinking about my project was about increasing the accuracy by dividing down the diameter of the indexing wheel. If I started with say a 2 ft diameter disc, properly bearinged of course, and divided the circumference into as many divisions as I wanted (360 in this case) the accuracy would improve the closer to the central axis you got. The task would be to accurately mark out or divide the circumference. One of my first attemps at this, sucessful also, was to use a loop of bicycle chain. The disc was about 16 inches in diameter, and I used it to make a sprocket about 10 inches diameter. It works very smoothly. My problem in this case was that 360 links would have required a disc almost five feet across to do it.

        With this table I intend to mark out a few common divisions on the circumference of the disc to save me having to count every thread each time I move it. Marks at 5 degree increments would probably be fine, and maybe a circle of 7.5 degree increments, needed for division into 48 parts. I'll get a lot of capability just from those two layouts.

        I could have made a larger disc, one about 12 inches diameter. That would allow 720 divisions of a circle using 1/4-20 all thread. The rod would need to be 36 inches long, so a starting point would need a longer rod than that to get a full 360 with the proper curve throughout. In hindsight I should have done this, because the size of this disc in not a limitation in any way on the milling table. A cutter could still make it to the center of the disc without the disc interfering with the column of the machine.

        By the way, to roll the all-thread, I experimented with the size of a disc to form it around. I started with a disc 4 inches diameter, but it rolled too tightly. The proper size disc to use is 4 3/8 inches diameter. I clamped the disc in the vise, trapping one end of the rod under the edge of the disc, then just carefully wrapped the rod around the disc as far as I could go, then reclamped and continued bending. I kept as much tension on the rod as I could while forming it around the disc, and I tried to keep the tension and the speed of bending constant. I'm not using the circumference of this loop to index with, I'm using the side. In any event, there doesn't seem to be any marking on the circumference from this process.

        Another idea- whatever part of the loop one would use to index with could have the threads filled with a dark epoxy, or even just paint. Then sand or file around that to expose the shiny metal of the threads. That should allow a TR opto device to pick off the signal and count the rotation without contact. Might be a good way to adapt to cnc.

        A final note- I used too much epoxy in the groove, so I'm having to scrape some of it out of the threads. Just a few spots, no big deal. I'll get a pic or two in soon.
        I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-

        Comment

        Working...
        X