A 3D-Printed Bowl Feeder For Tiny SMD Parts

[Andrzej Laczewski] has something big in mind for small parts, specifically SMD resistors and capacitors. He’s not talking much about that project, but from the prototype 3D-printed bowl feeder he built as part of it, we can guess that it’s going to be a pretty cool automation project.

Bowl feeders are common devices in industrial automation, used to take a big pile of parts like nuts and bolts and present them to a process one at a time, often with some sort of orientation step so that all the parts are the right way around. They accomplish this with a vibratory action through two axes, which [Andrzej] accomplishes with the 3D-printed ABS link arms supporting the bowl. The spring moment of the arms acts to twist the bowl slightly when it’s pulled down by a custom-wound electromagnet, such that the parts land in a slightly different place every time the bowl shifts. For the parts on the shallow ramp spiraling up the inside of the bowl, that means a single-file ride to the top. It’s interesting to see how changing the frequency of the signal sent to the coil impacts the feed; [Andrzej] used a function generator to find the sweet spot before settling on a dedicated circuit. Watch it in action below.

We’re really impressed with the engineering that went into this, even if we wonder what the vibration will do to the SMD components. Still, we can’t wait to see this in a finished project – perhaps it’ll be integrated like this Arduino-fied bowl feeder.

30 thoughts on “A 3D-Printed Bowl Feeder For Tiny SMD Parts

    1. I work in a plant that builds engines and we have tons of these. Generally you don’t flip them. You create a point where there’s a physical check to see if it is oriented correctly. If it is not, it will fall back to the bottom of the bowl.. (probably Flipping in the proccess). Then the next one hits that check and if oriented correctly will move past.

    2. as the poster says down below you don’t, so if you have a part that has equal gravity you are going to end up with upside down parts or rotated 180 parts

      upside down resistors are not desirable, but other 2 pin parts like diodes or even multipin IC’s and it becomes a problem. The only real use of bowl feeders in electronics is for mechanical parts like testpoint loops, or we have a custom made part that the operators were swapping big ass reels out every 10 min, but again, its mechanical part being pick n placed

    3. We have some machines where I work that feed parts down a chute from the bowl, then a camera takes a picture of each part and orients it accordingly using turn tables or flippers.

  1. That looks very very nice. A job well done.
    I had to look a few times to see whether or not the bowl was 3D printed. The confusion was mainly caused by the dot-like-pattern in the center of the bowl. I wonder why that pattern is there. Anyway, nice job.

    1. This strange pattern appeared because there weren’t enough top layers and I forgot to turn the cooling fan on.
      Thanks!
      I’m gald you like it :)
      Stay tuned for updates ,its only a small part from a much bigger project.

  2. I would try a Reflective Phototransistor sensor to detect the tops of ceramic parts, and a small solenoid to bump items back into the feeder without the correct orientation.

    For some high frequency RF applications, people will sometimes mount metal-film resister parts upside down to make them more coplanar.

    Very cool project =)

    1. I’m planning to use some slotted optocouplers for part detection and counting. Actually there will be some “custom made” small solenoids in other parts of this project. Soon I will publish a article about it :).
      I’m glad you like it.

    1. Yeah, right? I love to see industrial hacks brought down to the individual level. It’s not like a PCB fab doesn’t have all the same problems that I do making PCBs at home — they just have much more interesting solutions.

      (Pun unintended, left anyway. Tee hee!)

    1. It might look fancy, but at the speed of professional pick and place operations these shakers would have to run at speeds that would just knock the parts out of the bowl. Our EMS runs machines that can spit down almost 20 components/second on the board (Siplace TX2, 100000/h theoretical, 65000/h per IPC benchmark).
      Good luck keeping up with that without a large reel.

      1. For small parts, there is a new cartridge format that does away with reels, and saves a lot of waste. This format requires some sort of vibration feeder that feeds the loose parts out of the cartridge, and onto a plate so that they can be picked up. Here is an example of how this is done: https://www.youtube.com/watch?v=EFP0RqktaR4. So this type of feeder is likely going be required at some point.

      2. For the hobbyist, speed is not a major concern. Stocking a wide variety of discrete components in pill bottles is easier than stacks of reels or tubes. If one really needed to up the dispense rate you could interleave multiple shakers to achieve 2x, 3x, 4x … depending on how complex you wanted to get.

    1. Gotta love how when someone bothers to document and show something off, someones gonna always try to take them down a peg, it was perfectly watchable and helped show off how it works.

    2. You obviously looked at something different than I did. Perfectly fine video actually showing the working of the device.

      The link to your videos showing how it should be done is missing BTW.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.