4 ms·

Brilliant idea, outsource to the human the low-precision positioning task and get rid of the workpiece fixtures for massive price reduction. Now someone needs

by 21 9y ago

Brilliant idea, outsource to the human the low-precision positioning task and get rid of the workpiece fixtures for massive price reduction. Now someone needs to build the equivalent for 3-5 axis, maybe with a free movable support arm since it might be too heavy to hold just with hands.

LeifCarrotson 9y ago
Honestly,I wonder if this concept isn't better suited for welding robots and plasma cutters. It seems ideally suited for those tools. You need precise control of a tool position, and the reason,they're not done by hand more right now is that access to the workpiece and clamping it in place are a huge pain. A big problem with handheld CNC is that cutting many materials you'd want to cut requires precise control of your speed, and also requires rather large amounts of force and rigidity. CNC in wood or plastic is pretty easy, but steel is hugely more difficult. There's a reason traditional mills are made with thousands of pounds of cast steel, and it's not to make it hard for hobbyists to enter, I wonder if a 3-5 axis might be better done by using servo-actuated brakes to constrain your motion to a desired direction and speed. Though I remain skeptical that the problem with CNC is the expense of long axes. You pay for the power, not length - a 30" ballscrew actuator costs maybe 2x as much as a 3".
- 21 9y ago
  You seem to know way more about the fields so I defer, but I think there is a market for just cutting wood, plastic or soft metals like aluminum. A low cost machine obviously will have a bunch of limitations.
  - philipov 9y ago
    2400$ isn't exactly low-cost.
    - icegreentea2 9y ago
      It's definitely not cheap, but its in the ball park for CNC wood routers (obviously you can find cheaper stuff, especially if second hand). There's value to some hobby users in that it looks like you get to save on bench space. For some users that might be the difference between having and not having a CNC wood router.
      - matte_black 9y ago
        Why wouldn’t a $1400 dollar shapeoko or x-carve be better than this?
        opencl 9y ago
        This doesn't take up an enormous amount of shop space, can be used on arbitrarily large workpieces, and probably requires far less calibration and maintenance work. I have a 2' x 2' CNC router and can easily think of usecases this would be better for. A 4' x 8' CNC router is sure as hell not going to be cheaper than this thing.
        froindt 9y ago
        I wonder how accurately it would work on a 4' x 8' piece. If I cut some curvy shape near the edge of the piece and followed the dashed line perfectly, would I end up exactly back where I started or would I miss by half an inch?
        hexane360 9y ago
        It uses strips of QR codes to position itself on the workpiece. It's an absolute measurement not a relative one, so there's no drift.
- linsomniac 9y ago
  The benefit I see is not the expense of doing long axes, but the problem with storing them. I can put a Shaper Origin away in a cupboard in the garage when I'm done, the 4x8 CNC at my makerspace takes up around as much space as a car, and hasn't been moved in years.

by 21 9y ago

LeifCarrotson 9y ago

Honestly,I wonder if this concept isn't better suited for welding robots and plasma cutters. It seems ideally suited for those tools. You need precise control of a tool position, and the reason,they're not done by hand more right now is that access to the workpiece and clamping it in place are a huge pain. A big problem with handheld CNC is that cutting many materials you'd want to cut requires precise control of your speed, and also requires rather large amounts of force and rigidity. CNC in wood or plastic is pretty easy, but steel is hugely more difficult. There's a reason traditional mills are made with thousands of pounds of cast steel, and it's not to make it hard for hobbyists to enter, I wonder if a 3-5 axis might be better done by using servo-actuated brakes to constrain your motion to a desired direction and speed. Though I remain skeptical that the problem with CNC is the expense of long axes. You pay for the power, not length - a 30" ballscrew actuator costs maybe 2x as much as a 3".

21 9y ago
You seem to know way more about the fields so I defer, but I think there is a market for just cutting wood, plastic or soft metals like aluminum. A low cost machine obviously will have a bunch of limitations.
- philipov 9y ago
  2400$ isn't exactly low-cost.
  - icegreentea2 9y ago
    It's definitely not cheap, but its in the ball park for CNC wood routers (obviously you can find cheaper stuff, especially if second hand). There's value to some hobby users in that it looks like you get to save on bench space. For some users that might be the difference between having and not having a CNC wood router.
    - matte_black 9y ago
      Why wouldn’t a $1400 dollar shapeoko or x-carve be better than this?
      - opencl 9y ago
        This doesn't take up an enormous amount of shop space, can be used on arbitrarily large workpieces, and probably requires far less calibration and maintenance work. I have a 2' x 2' CNC router and can easily think of usecases this would be better for. A 4' x 8' CNC router is sure as hell not going to be cheaper than this thing.
        froindt 9y ago
        I wonder how accurately it would work on a 4' x 8' piece. If I cut some curvy shape near the edge of the piece and followed the dashed line perfectly, would I end up exactly back where I started or would I miss by half an inch?
        hexane360 9y ago
        It uses strips of QR codes to position itself on the workpiece. It's an absolute measurement not a relative one, so there's no drift.
linsomniac 9y ago
The benefit I see is not the expense of doing long axes, but the problem with storing them. I can put a Shaper Origin away in a cupboard in the garage when I'm done, the 4x8 CNC at my makerspace takes up around as much space as a car, and hasn't been moved in years.