Week 16 - Wildcard Week

This week we have the following task to complete:

• Design and produce something with a digital process (incorporating computer-aided design and manufacturing) not covered in another assignment, documenting the requirements that your assignment meets, and including everything necessary to reproduce it

4th Axis Milling

With the introduction of electronics production and the "Make Something Big" assignment, we already learned how to use a CNC mill — but in my case, only for 2.5D milling. 4-axis milling goes one and a half dimensions beyond what we previously did.
The first project this week was to mill a bust of myself using the 3D scan from Week 05. I used both parts from the SLA print that were designed to house a USB stick, which fit this purpose perfectly. I reassembled the two parts in Blender with Ferdi’s help, as my initial idea of simply joining the parts and using the fill command didn’t work at all. In the end, I used the remesh function instead. Before this, I had already removed the internal geometry.

The next step was to scale the model to the correct size and import it into Fusion 360, which offers a powerful CAM environment for generating G-code for multi-axis machining. I had previously used Fusion 360 for 4th axis milling on the Carvera, with excellent results.

Zeroing the machine was slightly more complicated this time. In theory, the machine knows the position of the 4th axis and can determine the Z-height using the wireless probe. However, the probe is currently broken, so I had to estimate the Z-position manually. I also knew the Z-axis would be slightly off due to flattening the MDF wasteboard earlier, which removed a few tenths of a millimeter. To compensate for this, I replaced the wasteboard entirely and avoided using the full dimensions of the stock material.

Before starting the milling process, I prepared my stock by cutting a piece of scrap PVC we had lying around, hoping it would provide a good comparison to the SLA-printed version.

G-Code Programming

I used the following 4 operations for the milling process:

1. Rotary Pocket – roughing pass of the geometry using a 3,175 mm end mill
2. Rotary Pocket – creating a support cylinder at the back of the bust using a 3,175 mm end mill
3. Rotary Pocket – creating a support cylinder at the front of the bust using a 3,175 mm end mill
4. Rotary Parallel – finishing pass using a 0,2 mm V-bit

Modified Settings:
2D Pocket:

Rotary Pocket1:

Rotary Pocket2:
Same settings beside

Rotary Pocket3:
Same settings besides

Rotary Parallel1:

Machine Preparation

To install the 4th axis:

1. Connect the cable first.
2. Insert the dowel pins.
3. Mount the 4th axis and fasten it using six screws, tightening them in a cross pattern.

I also had to reverse the orientation of the chuck jaws to ensure proper clamping. Make sure to install the jaws in the correct order (from slot 4 to 1); otherwise, they won’t align concentrically, creating an eccentric rotation.

Milling Process

I broke the end mill because I had set the wrong stock width in Fusion, which led to an incorrect depth of cut—over 10 mm instead of the intended 1,5 mm. This caused excessive shearing force and ultimately broke the tool.

I suspected the issue was related to the Cutting Moves Approach Beyond Front Limit option, so I disabled it. I also limited the toolpath to only include Rotary Pocket 2 to validate the setup. After that worked correctly, I planned to run Rotary Pocket 3 and finally the Rotary Parallel operation.

Rotary Pocket 2 worked well, but during Rotary Pocket 3, the material shifted slightly due to torque from the 4th axis motor. Even after halving the step-down in the G-code, the problem persisted.

Since this part of the geometry wasn’t critical, I chose to skip it and proceed to the finishing pass.

However, this decision had consequences: the tool’s silicone collet would collided slightly with the uncut stock, preventing a clean finish.

After milling, I carefully cut out the part and sanded the base to make it flat and stable.

Outcome

The result was a major failure. I’m still unsure where I went wrong with the 4th axis sculpting, and I no longer have time to iterate.

I’ll present the result as a creative interpretation.

Electroplating

No other assignment comes close to the complexity and potential outcomes we explored and achieved through electroplating.
The goal was to electroplate an FFF-printed funnel for a friend who uses it to pour dietary supplements into flasks for trail running. Although PLA is food-safe, frequent use and hand-cleaning caused delamination over time. To make it more durable, I reprinted the funnel and coated it with graphite spray.

The first version had no hole for suspension in the electrolyte bath, so I printed a second version with a small hanging tab to facilitate immersion.

I performed the process with Jarni and Richard. Each of us electroplated our own part, but we shared the electroplating facility.

Preparation

To calculate the required current, the surface area of the object must be known. In my case, the funnel had a surface area of 7,984.83 mm².

To make the surface conductive, I applied two thin layers of graphite spray, with 10 minutes of drying time between layers, followed by 24 hours of complete drying. I used copper wire through the hanging hole of the second funnel and wrapped it through the center of the first, which lacked a side hole.

Before electroplating our actual parts, we tested the process on a sample piece with the following results:

During the initial electroplating attempt, the coating did not adhere to the parts. To solve this, we polished the surfaces using a paper towel and a polishing machine in the university's electrochemistry lab.

Thank you to the team for your support and for granting us access to your facilities.

Before starting the electroplating of our actual parts, we used a test piece to ensure that the plating facility was functioning correctly.

Electroplating Process

1. Submerge the part in HSO Superclean (55 °C) for 5 minutes to remove grease.
   
2. Rinse briefly in distilled water.
   
3. Submerge in 10% sulfuric acid for 30–45 seconds for surface activation.
   
4. Rinse again in distilled water (using a different container).
5. Begin copper electroplating using Cubrac 480 (contains copper sulfate, copper, sulfuric acid, sodium chloride, and organic additives) at 2.1 A. Initially, I aligned the funnel's hole with the electric field, near the air injection, but the outer edges received too much plating (even starting to "burn"), while the interior remained mostly uncoated. I later switched to a smaller electrode placed inside the funnel after achieving good exterior coverage.
   
   During electroplating of the test piece, we also observed a burned edge. The resulting surface variation remains clearly visible:
   
6. Rinse in distilled water. At this stage, the part is copper-plated.
7. For additional coatings like nickel or chrome, a copper underlayer is always required for better surface smoothness and corrosion resistance.
   Chrome, while corrosion-resistant, tends to form microcracks. I decided to plate one funnel with copper only and the other with nickel using ELPELYT GS 6 (contains nickel, chloride, and boric acid).
   This bath was held at 55 °C, starting at 1 A and increasing to 2.3 A. Starting with lower current improves adhesion; higher current increases plating speed and thickness but also raises the risk of “burning.”
   
   To achieve a more homogeneous surface on complex geometries, all plating baths are equipped with the capability to move the parts continuously back and forth by connecting the holding frame or rod to a linear actuator mounted at the rear of the setup.

“Burning” here does not refer to combustion. It’s a term used by the chemist to describe surface defects caused by excessive current.

The test piece has two different sides of roughness one side is smoother then the other, which allows for a good comparison between different surface finishes, which looks like that:

The result was a success and I only had to remove the wire that we used to create a small but good connection to the part:

What I Learned This Week?

• If the stl that you import into Fusion have some artifacts left it will create a warning "triangle group not found".
• The process of electroplating and all the steps it requires. I thought it would be easier and more complex at the same time.

What I want to improve next week

• Document while doing and not just taking pictures and taking some notes.

Design Files

bene
funnel

To create this page, I used ChatGPT to check my syntax and grammar.

Copyright 2025 < Benedikt Feit > - Creative Commons Attribution Non Commercial

Source code hosted at gitlab.fabcloud.org