The Challenges a 3D Printer Faces Cutting its Printed Model’s Apron Strings
Wait…. What? I can hear your voices echo through the ether. What on earth is she talking about? This is now a step too far. How can she prescribe human emotions to a machine? How is a printer emotionally attached to the model it is printing?
Well… it is a 3D printer, a technology born during the Fourth Industrial Revolution where according to Klaus Schwab, the originator of the term, had this to say in his book of the same title:
“Previous industrial revolutions liberated humankind from animal power, made mass production possible, and brought digital capabilities to billions of people.”
“This Fourth Industrial Revolution is, however, fundamentally different. It is characterized by new technologies fusing the physical, digital, and biological worlds.”
Therefore, in my most discombobulated state, and considering this definition, it is reasonable to assume that my 3D printer has merged with a biological being that can display human-like emotions.
OK. That’s taking my printer’s misadventures too far. I can only say that the journey to 3D printing success has been — and still is — long and frustrating. Some days, it feels like my brain is confused from trying to work out how to solve the latest challenges.
And yes… I know that 3D printers cannot be emotionally attached to the models they print. But, in this case, it seems as though they are. After successfully printing a Japanese Lucky Cat, I tried printing a dog ball, and the printer went haywire again — continually banging and crashing along its x, y, and z-axis limits, trying to emulate the Great Glass Elevator mentioned in the article on printing a Japanese Lucky Cat.
Troubleshooting a Crazy Printer
As a starting point, I went back to the tech hospital’s trusty technician, who suggested I print a Benchy that he sliced. Of course, as you can imagine, the Benchy printed perfectly.
What now?
Upon further reflection, I asked him to slice one of my models — a cat ball for my daughter’s Bengal cat. He happily did so and emailed me a gcode file to try.
Can you guess whether the cat ball printed successfully or not?
Of course, it did.
But… what was very interesting is the “crazy” behavior started becoming apparent. Even though the printer finished printing the Benchy and completed the entire print cycle with all the correct completion messages, it still tried to continue printing the Benchy when I switched the printer on with the SD card containing the Benchy’s gcode file on it.
An error message showed that the printer hadn’t completed the print. Do I want to cancel or continue? And even though I canceled the print, it still navigated to the start of the print cycle again. Stopping the print did nothing else than confuse the printer’s mainboard, causing it to try and print more than one gcode file simultaneously. This, in turn, resulted in frantic behavior, where the printer’s print head and bed crashed into their respective x, y, and z-axis limits. When that wasn’t successful, the print head tried to fly off the printer’s top, bending the Bowden tube and cables running between the printer’s mainboard and the Micro Swiss Direct Drive Extruder.
The only solution, it would seem, is to ensure that the completed print’s gcode file is no longer on the SD card when starting a new print. Hence, the printer is facing serious challenges cutting its completed model’s apron strings.
The obvious question is: Why is the printer getting so confused?
I really don’t know.
I asked my trusty tech hospital technician. He said they have experienced such issues before, and the solution is to reformat the SD card. Well… I went one better and purchased a new SD card. I have also moved my 3D slicing software and models onto an old i3 laptop that I can access using Real VNC — a headless setup using one monitor, keyboard, and mouse. I haven’t let the new SD card near my main laptop — a MacBook Air. There are known issues — possibly only anecdotally — printing models sliced by software installed on any macOS device, especially the M1 and M2 devices.
Conclusion: The Firecrab Take
My mantra — and thread running — throughout these articles on my 3D printing adventures/misadventures is that my misadventures are mainly due to my inexperience with “all things tech” as well as the total lack of documentation aimed at troubleshooting straightforward issues, never mind seemingly obscure issues like the printer attempting to retain ties to a completed print.
A second challenge facing new users of technology is that product owners are relying on Generative AI and Large Language Models (such as ChatGPT) to write their technical documentation.
Don’t get me wrong. GenAI and its current iterations have produced some of the most impressive technologies I have ever seen — and there is no doubt that this technology will continue to improve over time. But at what point did product owners think removing human, technical writers from the equation was a good idea?
Sure, the role of technical writer might have changed from wordsmith to word orchestrator, but there still needs to be human oversight. Otherwise, like my 3D printer decided that trying to print past, present, and future models simultaneously was a good idea (with disastrous consequences), the latest LLM iteration might decide that formulating its own truth is a good idea with equally disastrous consequences.
