Hello, I just made an account here. Consider this my introduction:
I play many many instruments, but I’m not an expert at many of them. Recently I became interested in Irish trad music (on my mandolin), and playing at the local session renewed my interest in simple flutes. I have made several piccolo/fifes in the past, so I made some larger lower PVC flutes. I also just finished 3D printing the Marat Axianoff flute, which is so cool to play.
Now, I am interested in making some whistles…. I have a Clarke Celtic D, and a random plastic one, but making instruments is like potato chips, can’t just have one. Here’s what I’m wanting to do- I would appreciate feedback.
Since I have a 3D printer and some CAD experience, I’d like to make the mouthpiece this way. It’s precise, tweak-able, and most of all, repeatable. I want to make the rest out of metal tube (type K copper possibly). The catch is I want the labium (I think my terminology is right?) to be the edge of the metal pipe.
I created a quick mock-up to illustrate what I mean. See pictures. Please note that it’s not proportional nor to scale, and the edges would get finished much differently. Wondering what some more experienced people think.
I’m not an instrument maker, but I wanted to add a couple of thoughts to this discussion, before the experts chime in.
First, I’ve encountered all kinds of whistles, but I’ve never encountered one like this, where the labium is literally just the unmodified edge of the tube. Because I’ve never encountered this, my inclination is to say it wouldn’t work properly. On every whistle I’ve ever encountered, the labium tip is LOWER than it would be if you simply used the edge of the tube. So even whistle designs that use the edge of the tube as the labium will still modify the edge, usually by making it incline downwards, so that its tip is lower down than the tube itself. See, for example, the MK Midgie. I imagine doing this is necessary either to make the whistle play at all, or to voice it correctly.
Second (and less importantly), if you’re already 3D printing a whistle head, and the whistle is going to be cylindrical, you might as well add a reverse taper inside the head. This will fix the flat-second-octave tuning problem that most cylindrical whistles have. More whistle makers should do this, in my opinion. And it shouldn’t be difficult if you’re 3D printing the head. It would just be a matter of adding some extra material that goes inside the tube, past the labium.
This isn’t inherently unworkable, but there will be things to work out. Looking at the second image - the side-section - in particular, make sure that the floor of the windway is just a hair lower than the inside surface of the copper tube. You may want the roof of the windway to be higher than the outside surface of the tube, but I don’t think this will be needed.
Downstream of the window (to the right in the side-section), there is an exposed section of the copper tube, then a section of the printed head that wraps around the tube (light-blue hashed in the image). I would like to see that part of the head either level with the end of the tube, so no copper is exposed, or with a ramp on the leading edge, so there is no sudden step up from the outside surface of the tube.
In practice, I expect the tube will be thicker than your images suggest. You might be tempted to file a ramp in the end of the tube, so the leading edge of the “blade” isn’t a blunt wall. I don’t think this will be essential; you might want to try it both ways.
I’ve just checked my Generation whistles, and they’re very slightly flat in the second octave, but with my own quenas with whistle head adaptors they aren’t flat in the second octave, and that’s despite them widening slightly at the top end of the tube due to the internal ramp rather than narrowing. I don’t know why that happens, but with 3D printing you’d have the advantage that you can make lots of near-identical whistle heads with systematic variations between them to pin down the exact causes. The reason I have that widening is that it’s easier to blow the instrument in quena mode with the splitting edge of the labium further out than the bore, while whistles and recorders tend to have it dip further in. The proposed design at the top is half way in between these.
You will have to ensure the wind-way floor is absolutely smooth to create a steady laminar flow wind-sheet. 3D printing alone is not good enough for such a surface finish on the floor. Yo may want to bear in mind you will have to smooth the floor after printing. Good luck with the project
If I may quote my own review: “Lastly, a word about the labium ramp: There isn’t one. That “blade” is the end of the copper pipe where it was cut with pipe cutters or something. It hasn’t been sharpened, thinned, or worked in any way. I was quite surprised it made a sound at all after looking at it. But it works!”
To answer @Chip directly: 3d printing plastics are cheap and tubes are cheap. It doesn’t hurt anything to just give it a try and see how it works.
There’s a number of folks (myself included) making 3d printed whistles these days. It allows for pretty fast iteration for dialing in the geometry. And once it’s dialed in, you probably don’t need as much post processing as you might think. I grabbed this femur-shaped whistle off of a reddit post and printed it and it played straight off the print bed.
Thank you all so much for your responses! I really should have said in my first post why I initially thought to do this- I think it will be easier to finish the edge of the copper tube into the perfect shape and smoothness than it would be a 3d printed part. My 3d prints come out pretty smooth, but nothing like polished copper could be. The printed fipple/windway will just give me more control to tweak various dimensions, while only needing a couple pieces of copper tube with different edges to test them on.
I have given some thought to this as well. I think I’ll print this Fajardo wedge of sorts after I dial in the tone I want, as it will need adjustment too. A tuning slide will become necessary as well, as the wedge can mess up the first octave. I don’t understand any math behind it, and what I read online (Mostly Doug Tipple and a guy on youtube) is that it’s all trial and error.
In response to Tunborough, yes, I will clean up the printed design to hopefully not mess with the airflow. It would be nice to not have any post-processing to do, but I doubt I’ll be that lucky.
To Wanderer, your reviews of the other copper whistle were helpful. I am encouraged that the first one had a plain cut blown edge, maybe getting sound won’t be too bad after all. Cool bone whistle!
I bought some copper pipe tonight to get solid measurements from, hopefully I’ll get a good model to print in the coming days!
Hi, it’s me again, O’Brien, with a new handle. I’m still making a few whistles, but I haven’t started my website up again after it got compromised a few years ago. Relying on word of mouth and folks who find me on YouTube. I’m into my eighth decade, so time to slow down, I guess.
Regarding the question about using the cut end of a copper tube as a whistle blade: yes, it works if you use a plumber’s pipe cutter. The edge of the pipe is left with a slight camber and there is a tiny ridge on the inside edge. These two features produce the correct airflow. I discovered it by accident. I made a whistle without filing a ramp, planning to modify it until I got a decent sound. It worked the way it was. I just had to adjust the window dimensions.
I can be reached at davidpeterobrien at yahoo dot com
