Frederik Van Melle

Freelance Cinematography

The Unfolding Ball Box (very long post)

A few years ago I decided to try and create a box that opens using some kind of spring mechanism. The design needed to be able to store something small so the remaining space could be used to house the mechanism.  I always wanted to learn how to use 3D modeling software and this project seemed like a perfect opportunity to get started. Additionally, I jumped in the 3D printer hype and bought myself a second hand DaVinci 1.0 printer.

Thus started my journey of discovery in a strange world filled with smelly plastic and failure.

The DaVinci turned out to be a bad choice to get started in 3D printing. After 6 months of frustration and delight, I sold the machine to a dubious Maroccan merchant and invested the recovered money in an Indiegogo campaign called Easyarts: ARES. 6 months later, there was still no machine. The company was unable to keep up with the demand and decided to cease all communication. By this point I knew my money was gone and I decided to buy a delta bot kit. I'm an expert in taking things apart so I thought, why not put something together for a change? 

When I bought a kit from, I was hopeful that a fast and precise 3D printer was right around the corner. I learned the hard way that cheap 3D printer kits do not produce fast and precise 3D printers. Especially when you have no idea what you're doing and are blindly following poorly made instruction videos. 6 months later, I was ready to throw the machine out the window. I had encountered nearly every possible issue with my home built 3D printer and was, much to my surprise, able to resolve most of them. Nevertheless the printer never ran reliable and is now disassembled for spare parts.

Plot twist: the ARES arrived on the same day as the builda3dprinter kit. After 6 months of painful tinkering with the DaVinci 1.0, I could simultaneously build a 3D printer and get started with the ARES as my first reliable delta bot. Looking back, If the ARES hadn't arrived, I would probably have given up on 3D printing after my miserable experiences with the DaVinci and the Delta from builda3dprinter. Despite the lack of any customer service, the machine hardly ever seemed to need any servicing. It's a beautifully designed and reliable delta bot.

I knew that if I wanted to print my boxy contraption, I needed a bigger build platform.
When I saved up enough money, I bought finally bought the machine I wanted from day one:

The Rostock Max ( V3 by now )

Despite it being build by me, it's super speedy and precise. Right now, it's still my machine of choice and is serving me well in my quest for a ballsy box and useless figurines!

Below you'll find my attempts and designs ( Autodesk Inventor, Blender ) .

1. The bal was still square.

I thought it would be a good idea to try a regular box first and see how it goes from there.
The design was done in Blender and printed on the DaVinci 1.0.


2. The box became round.

Changing the shape from a rectangle to a sphere turned out to complicate every aspect of the design. New problems arose for the simplest parts and me not being an engineer wasn't helping. Luckily, there's enough inspiration to be found on the internet and these are the designs i found most helpful:

Flower box by msruggles
Mechanical flower by lizje88
L'Esprit des Cabinotiers

Below is my first attempt at a real ball box.

The yellow stuff is glue.

The yellow stuff is glue.

3. New printer, new attempts!

This one was designed in Autodesk Inventor. It took me quite a while before I got the hang of the software ( I still don't really ). The print is pretty small, too small to actually work but I wanted to test some basic ideas. This attempt was printed on the ARES.



4. Reworking the internal mechanism and wings

I will spare you the previous 4 designs and skip straight to my latest attempt.
There have been a lot of modifications on the mechanical design but few on the main aesthetics. This one has been printed on the Rostock Max and the ARES.


The whole is larger dan the previous photo and is just for testing purpose.

The whole is larger dan the previous photo and is just for testing purpose.


This is where the project is at right now. I have a lot more material I could post of the previous development, but this post is already too long! The project will undoubtedly continue to evolve as I manage to squeeze in time to work on it.