Tuners and Nuts

2025-Dec-28

In the last few days I’ve made two 3D printed parts for my instruments.

Tuner Clip

A plastic plate that attaches a small guitar clip to the tuning machine

source code | printable file

My father gave me an itty bitty tuner. It’s so cute! It’s also small enough that it could go on the back of the head, out of sight, and just stay there as a semi-permanent feature of the guitar.

Unfortunately, the heads on my tenor guitar and bouzouki are crowded. The included clip only really works at the far end of the head, where I normally put the capo, and the tuning machines occlude the tuner in that position unless I mount it on the front of the head.

The manufacturer makes a clever version with a plate you can mount using a tuning machine screw, but I didn’t want to buy a new tuner just to get a plastic plate. I modified somebody else’s model into the exact part I needed.

This keeps the tuner hidden at the back of head, stays out of the way of the tuning machines, packs into the gig bag without issue, and doesn’t contact the wall mount where I hang the instruments. The tuner is now effectively a part of the instrument.

The OpenSCAD source code is linked above in git. I don’t imagine this tuner will be manufactured past 2040, and the code is kind of long, so I’m not going to drop it inline here.

Nut

My plastic nut on a tenor guitar

source code | printable file

I screwed up the nut on my tenor guitar. I was being careful filing it, but a big chunk of bone came out. After that, the string was buzzing against the first fret.

I knocked the old nut out with a piece of wood and a hammer, then measured it with calipers: 30mm×9mm×5mm. The nut is a pretty simple shape: start with a 30×9×5mm cube, then remove any part that doesn’t intersect with a 30mm cylinder with a 9mm radius, for a bit of a curve. I decided to add some initial cutouts for strings, but kept them shallow enough that I could fine-tune the depth with my nut files.

I made one prototype with 15% infill. It looked fine, so I made a “final” version with 100% infill. The print layers are stacked to pancake between the head and fretboard: I didn’t want to risk tearing off a layer while tensioning a wound string or filing the nut, but it probably doesn’t matter.

My printer made a top layer that was about 0.3mm thicker than it should have been, which was actually just fine: it gave me the opportunity to sand the top layer to a smooth surface. Once it was down to 5mm thick, I was able to wedge the nut into the neck. The strings were going in the right places, it felt right, so I started filing.

PLA files much more uniformly than bone! I did much more filing on the plastic nut than I did on the bone nut, which I think is a better situation: I didn’t need to worry about accidentally grabbing a higher-density blob than the rest of the material, like I did with the bone nut.

It took about an hour’s worth of filing and setting the intonation properly, and now the instrument plays again, with precisely tuned intonation on the notes closest to the nut.

The part is still gray: I use gray plastic for all my prototypes, and the 100% infill prototype wound up being good enough for this task. Traditional nuts run from grayish to ivory in color, so I doubt any non-luthier will notice this slightly-grayer-than-normal nut.

Plastic vs other materials for nuts

PLA is not as dense as bone, and is going to absorb vibration more. This means you won’t have as much sustain, and probably translates to higher frequency overtones being attenuated. There are endless articles written about this, and a guy on YouTube has actually done some testing and confirmed the conventional wisdom. The best material wound up being metal, though, so most instruments are compromising sound for the luthier not swearing at you, anyway.

But we are taking about the neck here. I’m just not playing at a high enough level that I need to worry about whether I’m attenuating overtones through the neck on open strings. And I’m not playing the kind of music where I’d need an 8-second sustain on a plucked open string, anyway. So, for me, PLA is just fine. And it’s better than bone in that it’s easiry to work with, and I don’t have to store a bunch of weird size nuts in my house: I can just print a new nut whenever I need one.

Besides, as soon as you press a string down, the fret is the thing that’s transmitting vibrations into the neck. Your finger is going to absorb most everything after the fret, and the nut will get hardly anything at all. And frets are made out of—surprise, surprise—metal! My new opinion is that unless you have a metal nut, you’re in amateur territory, and I reserve the right to laugh at you being precious about nut material.

OpenSCAD source code

// Size of the nut
size = [30, 9, 5];

// Strings you're putting on it, in mils (aka thous)
strings = [11, 16, 30, 40];

difference() {
  $fn = 180; // 180 faces per complete circle is good enough for printing
  inch = 25.4; // mm per inch
  mil = inch / 1000; // mm per mil

  intersection() {
    cube(size);
    rotate([0, 90, 0]) cylinder(h=size.x, r=size.y);
  }

  spacing = size.x / len(strings);
  for (n = [0:len(strings)-1]) {
    d = strings[n] * mil;
    pos = [spacing * (n + 1/2), size.y, 0];
    translate(pos) rotate([5, 0, 0]) hull() {
      translate([0, 0, 0]) cylinder(h=size.z*2, d=d, center=true);
      translate([0, 10, 0]) cylinder(h=size.z*2, d=d, center=true);
    }
  }
}