I’m designing a case for a dew-point ventilator controller to be 3d printed. The controller is implemented using arduino on an esp32. The project is based off of the code and HW implementation by Make Magazine Germany: https://github.com/MakeMagazinDE/Taupunktluefter. When starting out I was thinking this would be an easy project but it turned out that especially the lid with its lip and groove design and the parts fixations were not that easy. I’m excited to finally print it.

The file is parametric to some extent and the main footprint is based off of a master sketch. Many parts were imported as step files from grabcad. I used FreeCAD 1.0-rc1 which works like a charm for many things. Next thing I would like to do is to use the new assembly workbench. What do you think?

Manual “Exploded view”,

Opaque view.

  • @namelivia
    link
    416 days ago

    This is very cool, what printer are you using and what material for printing?

    • @callccOP
      link
      216 days ago

      I’ll try on my good old ender 3 with PLA. Should do the trick.

  • Cris
    link
    English
    316 days ago

    That looks awesome! The name is super cute, I love it

    • @callccOP
      link
      216 days ago

      Many thanks! I figured out the name should be Dewie while typing out Dewpoin… on the case body. It was too obvious to not use :)

  • @madnificent
    link
    216 days ago

    Looks great! Are your files openly available?

    • @callccOP
      link
      215 days ago

      Thanks, not yet since there are quite a few things missing still unfortunately. Once I’ve completed it and successfully printed it, I’ll publish it somewhere.

      • @madnificent
        link
        215 days ago

        I’d love to see the FreeCAD files and possibly see the progress too. Looks like a good project to learn from.

        I’ve only seen my own (sometimes messy and slow) FreeCAD files and I also don’t get round to publishing anything either. Last time I made an enclosure I made a mess where it became slow because a lot was recalculated based on the position of the components and it had some curved surfaces. I do remember making clips (did not work well) and a ridge (that held up a bit better). Would use screws next time like you’re doing here.

        • @callccOP
          link
          215 days ago

          Ok, since you’re really interested, here’s my file: https://acloud11.zaclys.com/index.php/s/x2QGkqcCDNjzWe2 I started out by using a master sketch but got a bit lazy afterwards, so quite some things are positioned manually. I’m not sure it’s the best file to learn from but maybe it’s not terrible either.

          • @madnificent
            link
            215 days ago

            It’s very interesting to scroll through and I could easily find my way in it. It is also very fast. Thank you for sharing.

            I have the tendency to lock everything in place and fixate it based on some logic. For example: I would constrain Sketch010 for PocketGlandHoles so each of the holes is exactly the same distance. Lack of experience makes me want to have a parametric model so I can move things around until I make up my mind. Looking at this file however, it may be better to just model again when things change.

            I’d love to have another peek if you screw up with this approach or if you’d have a finalized version.

            • @callccOP
              link
              114 days ago

              Heh, that’s also interesting for me to hear. I usually also tend to try and constrain everything and make it as parametric as possible. Since this model became more complex than previous ones I took the liberty to be approximate eg. when lining up the glands and the holes. In the end i just want something printable as quickly as possible. I’ll probably never change the dimensions again. It’s therefore more time efficient to re model things if I ever need to adjust things.

              With the new assembly toolbox I might find an easier way to fit things like the glands and the holes together. Maybe by just tying a cutting object and the gland together and using booleans (which I find quite cumbersome to use).