Test #7
This test could be considered both a success and a failure. But I learned a lot from it. Firstly, the brightener in the solution is a consumable and doesn't last forever. Essentially, I don't really need the brightening agent itself (the part will be sanded and polished anyway), but I need it to quickly coat the graphite with copper and create a smooth, even surface without the sandpaper effect. Since many of the parts have very complex shapes, sanding all the surfaces is physically impossible. Therefore, the surface must already be smooth and even.
Without a brightener, the copper lays down very slowly and rough, and is also very uneven. This means that some parts may already be coated, while others haven't started yet, and that's probably the worst part. I also changed the concept for creating the support structure. I won't be using the one in the photo above.
Ultimately, due to the uneven start and the very long process without a brightener (I'll tell you which one in the next test), I stopped this test, and the results were as follows. The parts that didn't start to coat with copper began to deteriorate due to prolonged exposure to the solution, and these parts were rejected. Those that were coated with a thin layer didn't allow for proper sanding, and the plastic texture became apparent (plus, you need to prepare the plastic surface better). And even where the coating was quite good, there were still spots that closed up late, and due to the difference in layer thickness, it was impossible to determine where and to what thickness to sand. That's why there were bald spots on these parts. But a few details still turned out pretty well. I blackened them. I used BrassBlack diluted 1:3 with water and exposed for 15 minutes, removing the film with a brush as I went. Then I rinsed thoroughly, dried for a long time, and removed the film with a toothbrush.
Overall, I was pleased with the result. The detail, color, shine, and texture of the metal are all very good. I've now ordered some gloss enhancer and copper varnish, and I think the next test will be even more interesting and as close as possible to the result I'm trying to achieve.
Test #8
This test was supposed to be the final one, but due to my mistake, everything went awry. The culprit (this is a preliminary conclusion) was the negative terminal. More on that later, but for now, let's take it one step at a time.
As I mentioned, I replaced the motor's power supply with a 3-volt one. This allowed it to rotate at 2 RPM (instead of 3.3). Because of the high rotation speed, fluid turbulence was causing grooves to form in the parts with holes, which shouldn't happen at low RPMs. I soldered two plugs to this unit, but reversed their polarity. Now, when I plug one plug into the connector, the motor rotates in one direction, and when I plug the other plug in, it rotates in the other direction. I couldn't think of any other quick way to reverse the motor.
I also bought copper varnish... and here I encountered the expected catch. The layer thickness of graphite varnish is 5 microns. But copper varnish is 10 times thicker...
I coated a test sample and realized that this varnish wasn't suitable for our purposes. So, after applying the graphite layer, a day later, I brushed on stripes of copper varnish. This creates areas of greater conductivity so that the parts are coated with copper directly along these stripes, and then the copper will cover the graphite faster, not just from the growth zone, but from the entire strip of copper varnish. That was the idea.
This time, I placed the parts not individually, but on shared holders, which is very convenient. And by the way, to accurately calculate the surface area of each part, you can find this information in Blender, allowing you to set the exact amp value instead of guessing based on the dimensions.
I also soldered all the terminals to a single brass wire, as they were regularly falling off the steel ring. I also drilled a hole on the other side of the plate so I could adjust its position for uniform flow.
And now, the most interesting part. Brightening agent. This is a component in the liquid that evens out the copper layer, makes it smooth (shiny), accelerates the copper plating process, and makes it predictable and expected. BUT! This substance must be used in a specific quantity! Exceeding the dosage will lead to dire consequences, ruining both the batch and the solution.
Thiocarbamide CH₄N₂S / SC(NH₂)₂ Thiourea.
Different sources provide different information on how much is needed for one liter. Some say 0.04 g, somewhere 0.06. But since I didn't know how much was left in the solution and didn't want to take any risks, I added 0.02 g per 1.78-1.8 liters. To make it easier to add very small doses, I measured out 1 gram and dissolved it in 100 ml of hot distilled water. Thus, in my case 1 ml = 0.01 g of substance.
Add very slowly, drop by drop, stirring well. Then let it settle and completely diffuse.
Now about the errors! I used a bearing to transmit current to the rotation axis... and while the rotation was at 3.3 rpm, the current fluctuated slightly due to the balls turning in the bearing, but relatively little. But when I reduced the speed to 2x, the voltage surges became huge, not 0.1-0.2 A, but could fluctuate between 0.4 and 1.8 A... and I only noticed this at the very end... ...and this is what it led to.
The components on the upper levels, where the contacts were closest to the voltage surges, were all damaged. This wasn't immediately apparent.
But when I started processing the details, the top two levels fell apart in my hands.
The copper was so brittle that it seemed like dried sand crumbling away. At first, AI and I assumed it was due to Thiourea, or rather its high concentration (although, in fact, I think we should have added more, but that still needs to be tested). But after analyzing the entire situation, we hypothesized that it wasn't the cause, but rather the enormous voltage surges from the bearing's rotation. This was confirmed by the fact that Thiourea (the shine agent) is organic, and it sinks downwards under the force of gravity...
...and that fog at the bottom—that was most likely all the Thiocarbamide concentration. And here's my second mistake! Although the parts were rotating, it was at such a low speed that the liquid barely moved, and all the organic matter settled to the bottom. Most importantly, it saved all the lower parts.
It's a fact that the bottommost layer was slightly swollen from excess copper plating, but it had just the right texture. The second layer, however, remained shiny but had some minor coating defects. The middle layer also lost its shine but became rougher. BUT! All of these layers were made of very strong copper. It could bend without crumbling. This means that the copper's fragility isn't due to the additives, but only to current surges. Stirring the solution during the copper plating process is essential. Rotating the parts prevents them from being coated unevenly. Stirring the solution ensures that all layers are of uniform concentration. And of course, a stable current is required, without surges or interruptions.
That's why I replaced the bearing with a disc, which the entire surface of the bracket constantly rubs against, and it should transmit current without surges.
Now I want to show you the results. I'm not 100% satisfied yet, but given the potential, I understand what they could be. But even this result is quite good. I blackened the parts. I sanded some, and not others, to get a feel for how each surface would look. But conclusions have been drawn, and I hope the next test will finally be conclusive... and then it's just a matter of manufacturing all the parts using this algorithm...
But I understand, practise makes the master driller.
