3d printed whistles, take 2?

[Terry McGee]

Wow, great stuff, thanks Wanderer. About how long would that have taken to print?

[Wanderer]

Wow, great stuff, thanks Wanderer. About how long would that have taken to print?

On my printer, about 2 hours. On some of the more standard hobbyist machines (like an Ender 3), about 10 hours.

[Terry McGee]

Wow, they have come a long way, haven't they.

And I guess, at least in theory, that it should be possible to do tapers and flares and other such tomfoolery, although the maths might get a little frightening. Or am I making too much of that?

[Wanderer]

Wow, they have come a long way, haven't they.

And I guess, at least in theory, that it should be possible to do tapers and flares and other such tomfoolery, although the maths might get a little frightening. Or am I making too much of that?

Designing that stuff in engineering software is actually pretty simple. And if you set it up the right way, you could even iterate prototypes with small tweaks until you get exactly what you want.

That's what I did with my blade geometry. I set it up at an angle I thought was appropriate, and then I printed head after head, adjusting the angle of the blade a half-degree at a time, until I got something I liked.



Learning CAD software is a big learning curve, though
With this design, if I wanted to change the angle of the blade and reprint, I just change one number in the underlying support sketch, and the rest happens kind of automatically. You could do the same with tapers, bore perturbations, etc

[Terry McGee]

Wow, that is exciting, Wanderer. So for example, you could move from cylindrical to a mild taper, with the head a bit bigger than where you had started, and the foot a bit smaller. Or go the other way, insert a small reduction in bore at the top of the tube, if you felt that the thickness of the body wall was flattening upper notes more than you'd like.

I still feel we need to couple up computer modelling and 3D printing. That surely is the way of the future....

[Wanderer]

Wow, that is exciting, Wanderer. So for example, you could move from cylindrical to a mild taper, with the head a bit bigger than where you had started, and the foot a bit smaller. Or go the other way, insert a small reduction in bore at the top of the tube, if you felt that the thickness of the body wall was flattening upper notes more than you'd like.

That's exactly right. Let's say this was the schematic for the body instead--I could make a tapered bore by typing in 2 new numbers for the head and the foot, and the software would generate all the body in between at the thickness already provided. That'd require moving tone hole positions and sizes, which also would require changing some numbers--or clicking and dragging, but that's much less precise.

I could choose any place along the length to extrude in or cut away material to introduce a reduction or choke inside. Once everything was to my satisfaction, I just export it out, throw it on the printer, and have a prototype pretty quickly.

Now, I make that sound easy to do. And if you're a CAD expert, it is. I'm not an expert, so it usually takes me some time to get everything set up in a parametric way. But once it's set up, tweaking and fiddling is super simple.

I still feel we need to couple up computer modelling and 3D printing. That surely is the way of the future....

Just wait until they train an AI model to iterate instrument designs for optimal results.

[Moof]

It's amazing watching the whistles just appearing. I wonder if filament's available that would make Pride Week whistles? They'd be fun and might well sell!

[Wanderer]

It's amazing watching the whistles just appearing. I wonder if filament's available that would make Pride Week whistles? They'd be fun and might well sell!

With 3d printing moving more mainstream, you can find just about any color you're looking for. There's even a site set up to track them. Though even it doesn't have all the brands/colors available--they seem to be focused on single colors (which are easier to catalogue and sort) than gradient colors--of which there are a lot of options.

https://filamentcolors.xyz/library/sort/color/

[Wanderer]

Another video..I'm trying to record tunes that cover the range from bottom D to 2nd octave B.
https://youtu.be/uv0R8BDNXaM

And a good pic of the tuning slide in place, since I haven't posted one before.


About once a month, I vend 3d printed stuff at a monthly pop-up market here in town. Lots of people are interested in hearing about the whistle. I even sell one occasionally. But a happy accident that I noticed is that this geometry handles the wind very well. We were getting a cold front moving through, and the gusty cold air was blowing directly across the whistle. I don't know if it's the half-moon blade, or what, but these thing played through it all except the strongest 25-ish mph gusts (that's 40kph for those of you who use a more sensical measurement system). I was quite pleased at that, since I play outside often, and handling the wind is one of the things I try to highlight any time I'm talking about a whistle.

To be fair, I can't take credit for doing anything special to get that good wind handling, but I'm happy it's there.

[Wanderer]

This will probably be the last post I make in the thread about my own whistles specifically, though I'll be happy to continue to talk about 3d printing and whistles in general. I've opened a small preliminary shop on etsy at https://whistlerscreations.etsy.com/ and so want to abide by the commercial policies of the board.

I've made the executive decision that all my whistles will have a black head, no matter the color of the body. The black filament I use is a known variable, whereas different body colors have different additives and such to give them their various colors and looks. Not all of these additives make for a clean-cutting plastic, so I can't always get a good blade out of them. The toneholes are less tricky than the blade, and I have gotten a set of reamers that let me smooth things out if the deburring tool fails me on a particular filament. Plus, not every brand of specialty filament (especially the shiny multi-color variety) will provide a full list of additives in their MSDS, whereas I can feel good about people putting this black filament in their mouths as all of the materials are accounted for.

If I stick with the black head I know that I can always make a decent whistle, and swap out bodies for customer preference. As an example, I've made a whistle with the metallic filament that gave me so much trouble a few weeks ago--using a black head, of course.

https://youtu.be/NLyuAcYm6Zg Skip to 1:33 if you don't want to hear me talking about it--I basically say the same thing in that first 90 seconds that I've typed above

[Terry McGee]

Thanks again for taking us along for the ride, Wanderer, it's been really interesting.

Do get back to us if there are further insights and breakthroughs. We want to be the first to know here, not the last!

[Wanderer]

Thanks again for taking us along for the ride, Wanderer, it's been really interesting.

Do get back to us if there are further insights and breakthroughs. We want to be the first to know here, not the last!

My next big innovation in the process will be switching to a .2 mm nozzle (currently using a .4 still).

It may not sound like a big leap, but a .4mm nozzle can actually make .16mm layer heights pretty cleanly and that's what I print at. A .2mm nozzle can make then as low as .08mm. that's pretty fine! You can actually go a smaller (.01mm for the .2mm nozzle) but usually that requires much fine tuning to make sure everything turns out really cleanly. But, once you have the settings dialed in, it's very repeatable.

I have a .2mm nozzle sitting in a box, but I haven't put it in yet. Some filaments (especially shiny "silk" filaments and sparkle filaments) have additives that can cause smaller nozzles to clog, and since I use both of my printers for more than whistles, I'd need solid justification (ie: plenty of sales) to warrant getting a printer devoted to .2mm nozzle prints.

To highlight the resolution difference, this pic has a "benchy" (a little boat used to stress test printers) in both sizes. The red is .4, the white is .2. The rainbow bit in the background is a giant benchy printed with a .8 nozzle.

[Terry McGee]

Heh heh, so your version of the "fine grained timbers" we so like to work!

Does it have any impact on density? Interesting to compare the weight of a current whistle and one made with the finer nozzle.

[Wanderer]

Heh heh, so your version of the "fine grained timbers" we so like to work!

Does it have any impact on density? Interesting to compare the weight of a current whistle and one made with the finer nozzle.

It shouldn't have any material change in density. The melted plastic isn't going to be compressed--just dropped down in thinner layers. It'll just take more layers to lay out as much plastic as with a thicker nozzle. In fact, a lot of industrial guys move to a .6 nozzle, because fine details aren't as important as knocking out a lot of parts as fast as possible.

The biggest impact on results is fit and finish--things will look and feel smoother, and you'll be able to create thinner/smaller geometry cleanly. I'm already pretty happy with the overall feel and look at .16mm, but a smaller layer height will hopefully help with blade geometry.

The biggest impact on production is time. It will take twice as long, basically, to lay down the same amount of plastic. But since whistles are mostly hollow tubes, it's not a huge impact like it would be for say, the fidget dragons I also make and sell. Making a half dozen whistle bodies would go from like 7 to 14 hours. But a plate of fidget dragons already takes like 18 hours to print. Moving to the smaller nozzle means it'd take a day and a half.

That's probably the biggest reason I've sat on this .2 nozzle for a couple months and not installed it--I sell way more toys at my local popup markets, and with 2 printers, I just barely am able to restock inventory between them. What I really need is a 3rd printer that's solely dedicated to those specialty things (like whistle heads) that I want to use a .2 nozzle with. I just have a hard time justifying the cost at the moment.

[t4r3e2w1]

Wow, they have come a long way, haven't they.

And I guess, at least in theory, that it should be possible to do tapers and flares and other such tomfoolery, although the maths might get a little frightening. Or am I making too much of that?

Designing that stuff in engineering software is actually pretty simple. And if you set it up the right way, you could even iterate prototypes with small tweaks until you get exactly what you want.

That's what I did with my blade geometry. I set it up at an angle I thought was appropriate, and then I printed head after head, adjusting the angle of the blade a half-degree at a time, until I got something I liked.



Learning CAD software is a big learning curve, though
With this design, if I wanted to change the angle of the blade and reprint, I just change one number in the underlying support sketch, and the rest happens kind of automatically. You could do the same with tapers, bore perturbations, etc

Can Blender be used for this? I previously used it to create a 3D model, which I then handed off for printing. However, I'm not sure if the print was made directly from the blend file or if it was transferred to another softwaree.