Industrial ASMR: Mining an Mcor Iris 3D Printer for Sound

Ah, the Iris, the 3D printer that you desperately want to love, but that physics and reality make it difficult to do anything but hate.

At least it makes some neat sounds between throwing random error codes with no real translation. "Limit Exceeded" is the new "PC LOAD LETTER".

Of course, after reading this instructable, you can probably blame a good number of future failures of the printer on me and my abuse of it...

It's a subtractive 3D printer that uses paper. Like, plain ol' computer printer paper.

It does, well, what you would expect:

• Pulls out a piece of paper
• Cuts it up
• Puts glue on it
• Repeat for 10s or 100s of sheets, until finished or weird error.

To be fair, the weird errors happen because this is a VERY unstable environment to 3D print in/with. Since arbitrary paper can be used, you never know how much dust buildup you'll get in the machine while printing, whether the knife is out far enough to cut correctly, whether the glue wheel will jam, etc. I try to give Mcor the benefit of the doubt here because when things do come out without errors, they're really, really nice looking and sturdy. Not to mention, the waste material is recyclable.

The video attached to this step will give you an idea of what the process looks like, and why it's such a sonically interesting piece of equipment. There's so much going on inside of it between stepper motors, the glue wheel, and the slicing of the paper. So throwing mics inside of it should either be fun, or a complete mess, or both.

As you can see in the video in the last step, there's not a lot of working room in the printer itself. The knife/glue wheel head has a LOT of movement freedom, and doesn't leave much space for lodging mics into the work area.

So instead of trying to put everything on the print bed, why don't we just make the print bed a recorder?

I built a simple model in Inventor to embed contact mics in (for more information on recording equipment used, see the intro instructable to this series), then printed it on the Mcor. This can be attached to the bed of the Mcor so that the printer will lay paper down on top of it, and allow us to hear the knife and glue wheel doing their work up close.

Usually, the Mcor expects the outside of a print to be waste material, stripped off before extracting the rest of a print. Since we need the platform to take up the whole 8.5"x11" bed of the printer, I ended up cutting holes into the side material by hand, so we can run the cables for the mics out the side. Messy, but it did the job.

Now for the fun part, trying to get the platform into the machine. I taped the recording platform where paper would normally go, meaning when the machine does its calibration sequence (moving the platform up and down), it will register this as the start of the print.

The problem then was getting the cables out of the machine. This took a lot of testing with mounting the wires in different places using blue tape, to make sure there was enough slack for the platform to move, without the wires getting caught or in the way of the cut/glue head. I got lucky in the fact that the door latch still allows some room between the door and the machine (as seen in the images by me sticking my fingers under the door), meaning I could run the mic wiring out the front to my recorder.

Since I couldn't resist throwing more microphones in the machine, I also added one of the Sennheiser mkh8040 cardioids, mounted on a plate to the side of the paper loading mechanism. The print head comes VERY close to the mic, but never hits it. Once again, sheer luck, but this allows me to get more of the lovely motor sounds of the machine while it's working.

So what did it end up sounding like?

Unfortunately the platform wasn't quite thick enough to keep the contact mics from clipping when the cut/glue head went over them, and there wasn't much I could do in terms of compressing or readjusting gain. That said, they still work quite well most of the time, and the MKH8040 really added some nice detail for the motors in the machine.

The original idea was to do a full 4 hour print recording, so listeners could hear the differences as the paper model built on top of the platform. However, as of 2 hours in, there wasn't much in the way of difference in what was heard in the contact mics, so I gave up on that idea.