W2: Computer Aided Design

So... I have no experience whatsoever with 3D. In fact I've been avoiding it for most of my life, just like video editing, because it always seems SLOOOOOW compared to the lightness of a code editor or 2D UI tools like Figma.

Goals

Try different 3D software
Understand the basic mental model of 3D modelling vs. 2D grpahic design
Explore very basic operations like extruding shapes and doing holes
Begin to understand how I would approach one my final project pieces

At the begenning, I thought I'd seize the opportunity to approach some ideas for the [note: pre-pandemic] final project, as seen in this sketch.

Resources

I decided to check the following tools:

I really like Tinkercad, but it seems to be pretty limited. OnShape lured me with the promise of browser based editing, but in my experience it was a bit clunky.

Initially, I settled with Fusion 360 for the follwing reasons:

Does all the things that we seems to need
Free license for students and educators. Simple registration and set up process
Lots of documentation available

Speaking of documentation, here are the resources I checked:

Autodesk has a number of training videos in their platform once you log in, although I found them geared towards people who already know 3D and are transitioning to Fusion 360
Autodesk also has quite a few free training curses on Udemy.
The book "Fusion360 for Makers", from Make Media really starts from the bare bottom.

Explorations

So I started with the basics... drawing a shape and trying to punch a hole through it. It took a while to understand it, because the process' implementation is very different from 2D boolean operations, despite it being conceptually equal.

In this case I'm drawing another "sketch" to create a new piece of the pedalboard.

It also gave me the opportunity to explore constraints, like parallel lines.

This is how both of them look together.

A very simple render in wood:

Final Project Designs

The machine box

As mentioned on Principles and Practices, due to the COVID pandemic, I changed my ideas for a Final Project, so the previous exercises are somewhat disconnected with the following.

This design was machined on flat 10mm plywood, so I worked in 2D in AutoCAD:

The spinning frame

As usual, I had to go through several iterations until I came up with a design that worked as expected. Early on I went with a circle. Here are a couple of designs for that:

Later on I decided it was a better fit to go with a rounded-corner rectangle, which I tried in 10 mm but ended up having to go all the way up to 20 mm thickness:

The gears

I designed the gears with the online tool Gear Generator

Then imported the result in AutoCAD to add a cut for the servo's fixture:

The PCB box

For the Hello Fab Kit board I made a simple black plexiglass box with openings for the connectors, LEDs and switches:

The control panel

This is really a very simple piece with placeholder holes for all the ports and switches in the control panel, to be laser cut in plexiglass. I made this in Gravit Designer, sort of an Illustrator on the cloud.

The 3D Printed Fixture

For this part of the final project I decided to give Shapr 3D for iPad a go. I wanted to try something that felt closer to designing on paper and I have to say that it has worked very well for me, and has been so far my favourite 3D app. I'm sure it can't handle the same level of complexity as other industry standard tools, but for relatively simple designs it has been quite a joy.

Quite a bit of my final project designs could be done in 2D, as seen previously and in the Computer Controlled Machining assignment. But I also needed to design a key part to be 3D printed. Here's the sketch:

There were, of course, a few variations and iterations over the orginal idea depicted in the sketch above: two and four screws, with and without the final prong, etc. This was the first one:

And this is the final one:

Learnings

Obviously this is not a finished piece, but it has all the necessary elements already. I need to polish and create the rest of the pieces following the same approach.

I've got to say that it was —maybe unsurprisingly— way harder than I anticipated, mostly because 3D and 2D software have such different interaction and conceptual models, even the utterly basic stuff like panning and zooming, or even copy/paste. This is by far the most frustrating part, but I guess also the point of the whole exercise :)