3d printing

So, some backstory: I used to make whistles. I made them out of exotic hardwoods and also Corian (that stuff countertops are made out of). Link is to a 12 year old thread where I was first advertising the Corian whistles.
http://forums.chiffandfipple.com/viewtopic.php?t=46741

To make whistles by hand, you need:
  • A long bed wood lathe. Used to rough a wood blank into a cylinder, and to hold the drill for drilling the long bore. I got a crap one from harbor freight for about $400.
  • a self-centering implement to drill the bore, which has minimum wander. Reamers and gun drills are the implements of choice. I used a gun drill, and it cost me $300 for the drill bit.
  • An air compressor. Gun drills are hollow, and you blow something down them to keep them cool and clear chips. When drilling metal (for guns) a coolant fluid is used. For wood, you use air. This set me back about $59. But it barely did the job. If I were to do it again, I'd get a $200 model.
  • A machine lathe, because precision counts if you want your musical instruments to be in tune. I got a mini lathe from china for about $800.
  • lathe tool bits and other parts such as shims. (this stuff ain't cheap yo). I probably spent $100 here.
  • A drill press and associated parts (jigs for holding a cylinder, drill bits, etc). Probably spent $250 here.
  • Various parts and bits: micrometers, calipers, breathing masks, goggles, etc.
  • A table saw for cutting wood and corian down to 12x1x1 size. Spent about $600.
Throw in workbenches, safety equipment, etc, and I probably spent $3000 getting the bare necessities. Not counting the price of the materials to make the whistles. Knowing what I know now, I'd made some definite changes and the cost would be about $6000 to get started. I sold all this stuff when moving to Virginia.

That said, I don't wanna spend $6000. Nor do I wanna clear out my garage and fill it full of wood and plastic chips that get everywhere. Nor do I want to spend 4 to 6 hours of labor making a whistle which might (due to unseen defects in the wood grain) catastrophically explode and fly apart on the lathe. Thus, my theory: 3d print the things.

I took my $1000 O'Riordan whistle and my micrometer and made a bunch of design diagrams.
oriordan head specs 2.png


Got on Fiverr and started looking for someone to turn those into a shape file for 3d printing. Last night, a guy offered to do it for $5.00. Most of the quotes I got were between $50 (for developing countries) and $360 (a mechanical engineer from Germany). Thought "what the hell? It's only $5.00. If it doesn't work out, I can still get one of the other guys to do it."

Woke up this morning with this in my mailbox:
wistel v2.png

Threw the .STL files in a viewer, and they look good, with good dimensions, so, I sent them off to Shapeways to have a prototype made, which should arrive first of July.

Total cost so far: $55. Kinda excited :D
 
Last edited:
What's the minimum resolution of the print job? It seems that if you're going to go 3D you should be able to do some experimentation with internal shaping and reflection that would be exceedingly difficult on a lathe.

--Patrick
 
What's the minimum resolution of the print job? It seems that if you're going to go 3D you should be able to do some experimentation with internal shaping and reflection that would be exceedingly difficult on a lathe.

--Patrick
I've given that some thought myself.
Cylindrical whistles have tuning trade-offs you have to make because of the nature of the standing wave in a cylindrical tube. If you were able to make a perturbed bore, you could eliminate those trade-offs. Bagpiper's do something similar by putting a wire with a bit of wax up in their chanters to change the tuning of specific holes.

It's definitely something I'm going to look into in the future. I plan on having these things SLS printed (laser melting a powder) rather than FDM (melted plastic extruduing out a nozzel), so my resolution will be a bit higher. SLS can theoretically go as low as .06mm for the layer height. With FDM the theoretical minimum layer height is .17mm. the .06 should be more than sufficient for my needs.
 
I've given that some thought myself.
Cylindrical whistles have tuning trade-offs you have to make because of the nature of the standing wave in a cylindrical tube. If you were able to make a perturbed bore, you could eliminate those trade-offs. Bagpiper's do something similar by putting a wire with a bit of wax up in their chanters to change the tuning of specific holes.

It's definitely something I'm going to look into in the future. I plan on having these things SLS printed (laser melting a powder) rather than FDM (melted plastic extruduing out a nozzel), so my resolution will be a bit higher. SLS can theoretically go as low as .06mm for the layer height. With FDM the theoretical minimum layer height is .17mm. the .06 should be more than sufficient for my needs.
Willing to share the files? I have a resin printer here at work that has lots of downtime, and that they let us use for personal stuff as long as it's not nuts amounts of material used. It has really high resolution too.
 
Willing to share the files? I have a resin printer here at work that has lots of downtime, and that they let us use for personal stuff as long as it's not nuts amounts of material used. It has really high resolution too.
I'll shoot you a pm a little later once I've had my afternoon nap :D
 
I'll shoot you a pm a little later once I've had my afternoon nap :D
I have a friend here who is also godly with tolerances and such (it's literally one of the things she does as part of her job designing/prototyping parts), so if those two pieces are meant to fit together, she'll tell us if they will, and/or the right way to print them so that they will (with only a little sanding perhaps).
 
Talked with my friend, and she recommended you look at this: https://docplayer.net/23837746-Chap...imits-and-fits-tolerances-and-deviations.html

Table 6-1 describes the kind of fit you want between your pieces. Table 6-2 describes what tolerances you need. The quick guide is that everything is in millimeters, and find your diameters on the left-hand side. Take the max minus min for your nominal size, and compare that number to what the advertised resolution of your printer is.

For example, If it's around 15mm in diameter, and you wanted a close running fit, look at T15, and take shaft max minus min. That's 14.984 - 14.966 which means you'd need a tolerance of 0.018mm. If your tolerances are coarser than that then you may get something different than ordered.

But for producing only a few, there's always sanding. In that case, order tighter than you intend.
 
Got my Shapeways products in today. The tuning slide actually worked pretty well with the tight tolerances. It was a super-snug fit, but after putting it together and taking it apart a few times, it loosened up enough to be just right.

I suspected I would have to do some surgery on the blade of the mouthpiece to get the thing to play properly. And I was right.

But...



The thing does play :D
 

GasBandit

Staff member
Shapeways was the folks I got my minecraft 3d printing done through. They do good work, though sometimes their packing isn't the best. I had to repair a broken piece of my object when it got to me, using cyanoacrylate glue.
 
My pieces came in a box that was close to "medium priority mail box" and was 90% bubble wrap. You coulda probably jumped on it and it be ok.
 

GasBandit

Staff member
My pieces came in a box that was close to "medium priority mail box" and was 90% bubble wrap. You coulda probably jumped on it and it be ok.
Mine was bubble wrapped up a lot, but it had a thin part that stuck out up at the top like an antenna (or rather, a mining pick), which was not reinforced on the inside, and the bubble wrap ended up breaking it off. But like I said, a little super glue, and it went right back on no problem.
 
Since I have two heads, I took the 'worse' of the two, and did a little more experimentation today.

I changed the blade angle a little bit right at the tip, using some jewler's files that came today. Made the angle a bit steeper just right at the cutting edge. Fixed a lot of the flaws. It's still breathy sounding (which some people actually like), but it fixed some of the octave/tuning problems and fixed the octave break breathing pressure requirements.

Even if nothing else, 3d printing a bunch of these and experimenting will really increase my whistle construction knowledge.
 
3d printing is so much easier to prototype with than a lathe.
Asked the original STL file designer to make me 10 different variations of the head, each with a number embossed on it (so I can know which I like best) so I can have them printed up. That'd be so much work doing it all by hand
 
Last edited:
lol, I've already had a couple of people turn their nose up at the idea...but they're budding whistle makers--which can be an insular and snobbish group.

I swear, one of them told me (paraphrase) "there's no way you can measure everything with a micrometer"
I was like "Uh, yes, you can...except maybe the blade angle. But that's what reverse trig is for".

Then I posted the video of me playing the 3d printed whistle and he shut up ;)
 
If you have any spare bodies (and they're made of the right material), you might try fuming one with acetone to see how that affects the playability and the sound. Not sure it would work well on the head, since fuming might unacceptably change the blade geometry.

--Patrick
 
Top