Christmas Countdown

Brief

Christmas is my favorite holiday and I wanted a desktop countdown that I would actually like having on my desk. The premise was simple. Two cubes with numbers on their faces, set in a sled-shaped base, that I could rotate each day until they read 0 1, then 0 2, all the way through 2 5.

The puzzle inside the project was that two cubes only have twelve faces total, and you need to represent twenty-six distinct two-digit days (01 through 25) by flipping which cube shows the tens digit and which shows the ones digit.

Process

Inspiration and sketching. Looked at four reference styles, sled with cubes, vintage truck with cubes, house frame with cubes, gnome-themed. Picked the sled for its proportions and simplicity. Sketched the sled in pencil with rough dimensions before opening Onshape.

CAD. Modeled the sled in Onshape with a part studio and a dimensioned drawing. The first version had a long extended back to balance a Christmas tree decoration on, but without the tree the proportions felt off, so I shortened the back.

Prototyping cubes. Tried multiple materials for the cubes. Maple, walnut, and a 3D-printed PLA cube as a control. The PLA cubes had crisper numbers because they were printed to spec, but they did not have the feel I wanted on a wooden ornament. In my head, the cubes had to be wood.

The number puzzle. This is where the responsible AI use assignment came in. We were told to use AI for something specific and document what we asked and what we got back. I gave the model the constraint (two cubes, six faces each, need to display every number from 01 to 25) and asked it to work out which digits go on which cube. The arrangement that works is Cube 1 with 0, 1, 2, 3, 4, 5 and Cube 2 with 0, 1, 2, 6, 7, 8, where the 6 flips upside down to read as 9. I confirmed the arrangement works for every day from 01 through 25 by enumerating manually.

Fabrication pivot. I planned to make the sled base out of wood but I could not figure out a way to cut clean slots for the cubes with the tools I had. Instead of grinding on a method that did not work, I pivoted. 3D printed the sled base, kept the cubes in wood. The hybrid was visually richer than either pure wood or pure plastic.

Skills Built

• CAD plus fabrication integration. A real project usually crosses multiple materials and methods. Knowing how to design a part that gets 3D printed while another part of the same assembly is hand-cut wood is a skill that has come up in everything I’ve built since.
• Hand sketching as a planning step. I sketched this in pencil before any CAD. Future me knew exactly what I was modeling instead of figuring it out as I went.
• Project planning with a Gantt chart. Even at four weeks, the chart kept me from leaving the assembly for the last day like I did on the Wall of Colleges sign.
• Responsible AI use. Define the constraint, ask the model, verify the answer manually. Use AI to skip past tedious enumeration, not to skip past understanding.
• Knowing when to pivot. The wooden sled base was a worse outcome than the printed one given the tools I had. Recognizing that early saved the project.

What I would do differently

I would experiment with the Cameo cutter for cleaner number engraving on the wood cubes, and add a couple more layers of finish to deepen the wood color. Both small additions that would push the visual quality up another notch.