Chiff and Fipple Forums

I'm trying to figure out how to think about aspect ratio for conical bore flutes. A single AR for such a flute doesn't seem to make sense... is the *mean* bore diameter above a hole the relevant measure to use? I'm having trouble figuring out what else might make sense, but this is a relatively new area for me.

We get used to the idea that simple relationships will predict outcomes. Halve the length of a guitar string (eg by fretting the string at the 12 fret) and we go up an octave. We conveniently forget the second-order effects, eg the end stiffness of the strings, making them seem shorter than they really are, the increase in string tension caused by the act of fretting, and so on. Measure from the 12th fret back to the nut, then on to the bridge, and you'll notice that "half way" isn't actually half way. Your guitar maker has been shielding you from the scary truth!

I reckon flute acoustics makes astrophysics look like a doddle. Their whole game is being enacted in a vacuum, so there are few complicating issues other than a million other stars trying to exert some minor gravitational attraction. You can deal with that....

But the flute is a total physics battleground. Sure, we have the basic tenet that, if we double the length of the tube, the pitch will drop an octave. Except we have to factor in the effects of the tapers in the body but the cylindrical in the head. Plus the body thickness (ie hole depth) at embouchure and open holes. And the diameters of all these holes, never as large as they should be, and varying as they must to make up for the fact that if ideally placed, no human could reach them. And the helmholtz (cavity) resonance the stopper air gap represents, essential for getting the upper second and third octaves to behave, but complicating the rest of the spectrum. And the fact that, since no finger hole is large enough to represent a cutoff of the bore, we need to consider the stuff below that hole as well. Arghhh!

So, if we take any hole, we have to consider everything above it, and a fair amount (if not all) of what's below it. My own feeling is that all of that is just too much to take into account using just human mental resources. We have two options:
- continue with the time-honoured approach of snitching someone's earlier work - e.g. Messrs Nicholson, Rose, Siccama, Pratten, Firth & Pond - and reworking it to suit your needs, or
- embark on a cold-blooded investigation of the acoustics armed with new technology such as Wind Instrument Designer.

We've done brilliantly following the first approach, but I'm not seeing any massive leaps forward recently. (We'd easily recognise a massive leap forward. One maker would become fat and rich, while the rest of us would starve to death. It hasn't happened.)

I think we've reached the point where we have to start to stand on our own two feet, for the first time ever, since the invention of the tapered bore sometime back in 18th century France. Fortunately, I am very old, and excused such onerous challenges....

No, not even the venerable I am excused! I just have a bit on at the moment. A massive carillon upgrade, as well as the start of a new bushfire season requiring I bolster defences. (Malua Bay made it to the front page of the New York Times last New Year. We're hoping to keep a lower profile this year!)

Given the continuation of good health, and successful catching up after bushfires, covid, carillon upgrades, etc, I am looking forward to having a shot at using Wind Instrument Designer to realise some ideas I've been harbouring. It's going to take some getting used to. Computer modelling is an exercise in pernicketyness. The computer knows nothing, but needs to know everything. It will be tedious. But it will confront us with our place in the Total Perspective Vortex, as Douglas Adams so winningly imagined it. Infinitesimally small....

Malua Bay! We're practically neighbours. (I'm in Melbourne.)

Thanks for your reply. I am gradually coming around to the idea that the best method for improving my shakuhachi is going to be... learning a lot. Only somewhat sadly. I've written some programs to suggest hole locations, but one of the half-smart ideas I put in there is that every time I come across someone's simple model for hole placement, I put it in there and it's somewhat discouraging to see how different they are. Obviously there are a lot more variables at play beyond that.

For reference, I'm making instruments like these: https://thinger.org/tmp/2020/printed/. This is the latest to come off the printer, a 1.3 shakuhachi (one of the smaller standard sizes). Modifying the instruments post-production isn't really practical for me, so moving a hole a few mm is a non-starter, but I can adjust the model and print a new one. At this point, I try to make each print an experiment to learn something.

(The model itself is generated from a computer program, but the program doesn't know about acoustics: just how long the various sections are and where to put the holes.)

aplati wrote:Malua Bay! We're practically neighbours. (I'm in Melbourne.)

Thanks for your reply. I am gradually coming around to the idea that the best method for improving my shakuhachi is going to be... learning a lot.

Hello @aplati, from Melbourne!

You could contact David Brown here in Melbourne who is a highly experienced Shakuhchi maker of great excellence. Knowing David, he's always very helpful.

Andrew

aplati wrote:Malua Bay! We're practically neighbours. (I'm in Melbourne.)

Thanks for your reply. I am gradually coming around to the idea that the best method for improving my shakuhachi is going to be... learning a lot. Only somewhat sadly. I've written some programs to suggest hole locations, but one of the half-smart ideas I put in there is that every time I come across someone's simple model for hole placement, I put it in there and it's somewhat discouraging to see how different they are. Obviously there are a lot more variables at play beyond that.

For reference, I'm making instruments like these: https://thinger.org/tmp/2020/printed/. This is the latest to come off the printer, a 1.3 shakuhachi (one of the smaller standard sizes). Modifying the instruments post-production isn't really practical for me, so moving a hole a few mm is a non-starter, but I can adjust the model and print a new one. At this point, I try to make each print an experiment to learn something.

(The model itself is generated from a computer program, but the program doesn't know about acoustics: just how long the various sections are and where to put the holes.)

Those are unexpected! A very interesting concept. What is the notion behind the folded bore? A 1.3 shakuhachi is not especially large or unwieldy, and I must confess that the unusual bore and hole placement does look like it would trip me up! Or at least take some getting used to . I can see doing this with an ultra-low flute, for sure.

Andro wrote: You could contact David Brown here in Melbourne who is a highly experienced Shakuhchi maker of great excellence. Knowing David, he's always very helpful.
Andrew

Heh, David is a good friend of mine and my aikido teacher. He helped me make my first shakuhachi in his workshop, and has shared with me the bore measurements for several of his flutes, which I have used in my own.

Geoffrey Ellis wrote: Those are unexpected! A very interesting concept. What is the notion behind the folded bore? A 1.3 shakuhachi is not especially large or unwieldy, and I must confess that the unusual bore and hole placement does look like it would trip me up! Or at least take some getting used to . I can see doing this with an ultra-low flute, for sure.

A 1.3 shakuhachi isn't unwieldy, but now I can keep one of them in my pocket, and I could take an entire set of different sizes in a small bag without worrying about damaging them!

My original idea was to make more or less traditionally-shaped instruments, and I didn't know anything about 3D printing when I started. It turns out that printers on the less expensive side can't print an entire flute in one piece, even a 1.3, and it also turns out that the traditional joint style doesn't print very well because of the need for what they call "support," which is to say that if you try to put a thin line of plastic in the middle of nothing, it will fall down. So I started designing joints.

Before I got a joint that I was happy with, I got the somewhat ridiculous idea to try making a shakuhachi with a folded bore since that would eliminate the requirement for a printable joint. (I've since made pretty decent progress on joints, though.)

One of the major design considerations was that the fingering should be the same as traditional shakuhachi, so the unusual hole placement doesn't actually make that much of a difference in practice. I agree that without that, learning how to play them would be too much effort. One implication of the requirement is that a tube where the air is moving from bottom to top cannot have two holes for the same hand. Trying to make the ergonomics work out can result in some pretty bizarre shapes -- most notably the 1.8 linked above. But sometimes (particularly the 1.6 and 2.4) they end up being both ergonomically and aesthetically pleasing.

An unexpected (to me) side effect, as you noted, is that it would potentially make very low/long flutes playable. This aspect gets even better when I get to the point of designing flutes where the bore can twist in three dimensions, and in angles smaller than 180°: imagine that the bore could make a fold into a third dimension between two holes, so the bore distance between them could essentially be arbitrary.