PVC flute temperament

The Chiff & Fipple Irish Flute on-line community. Sideblown for your protection.
Tunborough
Posts: 1419
Joined: Sun Dec 05, 2010 2:59 pm
antispam: No
Please enter the next number in sequence: 10
Location: Southwestern Ontario

Re: PVC flute temperament

Post by Tunborough »

WIDesigner does not yet have an optimizer for setting the stopper position, but it does need one. Optimizing the bore profile of the headspace is probably not something needed for routine use, but it would be interesting to put one together just to see what happens.
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: PVC flute temperament

Post by Geoffrey Ellis »

Andro wrote:Hi Terry,

Curiously, I have been considering exactly the same thing about the headjoint for my design, having had a good look inside the Tipple headjoint with the wedge - just another bit of smaller diameter pipe shaved down to a wedge, with the inner curve still left in it.

Is the Boehm parabolic design ideal an actual parabolic curve, or just a conical approximation?
I've done extensive experimentation with both Fajardo wedges (like Doug uses) and parabolic tapers in the head, and there is definitely no comparison. The parabolic taper is far superior, if you have a way to pull it off. The wedges are certainly better than nothing if you are trying to get better intonation out of a cylindrical flute.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: PVC flute temperament

Post by Terry McGee »

Ah, thanks for that confirmation, Geoffrey. I've not done any extensive experiments there, I think maybe just one wedge inside one cylinder flute, and was quite put off. While the tuning was greatly improved, the performance seemed to drop way at about the same rate! Whereas I have made cylinder wooden flutes with a reamed Boehm-style bore in the head and got very satisfactory results. E.G:

Image

So, maybe take that on board, Andro. Something certainly needs to be done to overcome the cylinder flute's tuning, but the wedge is possibly not the best solution.

Now the other trick I've heard about but never tried is the Sanders Spike (from memory!). I understand it takes the form of a conical spike protruding from the face of the stopper and pointing down the bore. The cross-sectional area of the spike is designed to take up the same volume from the same places in the cylinder as the Boehm head taper would have reduced it. In practical terms, I imagine you'd make it part of the stopper arrangement. Because the air column is free to rotate about it, it might present less disruption compared to the Wedge (even though it might seem the opposite!). Can anyone report back on that?
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: PVC flute temperament

Post by Geoffrey Ellis »

Terry McGee wrote:Ah, thanks for that confirmation, Geoffrey. I've not done any extensive experiments there, I think maybe just one wedge inside one cylinder flute, and was quite put off. While the tuning was greatly improved, the performance seemed to drop way at about the same rate! Whereas I have made cylinder wooden flutes with a reamed Boehm-style bore in the head and got very satisfactory results. E.G:
That's a fine looking flute!

The spike definitely sounds like a better idea, and I'd be curious to try it. I don't love the idea of a spike in the headjoint for some reason, but conceptually it seems like it would cause less interference. Sounds like we had the identical experience, Terry: better tuning at the cost of a seriously choked off voice. They still played (after a fashion), but they were not free-blowing and they felt like they were fighting me. Some players loved them, but I never could.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: PVC flute temperament

Post by Terry McGee »

Geoffrey Ellis wrote:That's a fine looking flute!
Thank you. You'll note I did a few changes to the traditional (if relatively uncommon) 19th century cylinder bore flutes. Particularly at the joints, which are really thin-walled and often cracked for that reason. Especially as I would have cork-buffered the slides at the bottom of head/top of the body. So it's clearly possible to get around that issue without having to turn the whole thing into a baseball bat.
The spike definitely sounds like a better idea, and I'd be curious to try it. I don't love the idea of a spike in the headjoint for some reason, but conceptually it seems like it would cause less interference.
I think cleaning the head after playing will be a bit trickier, but probably not impossible. If you have a plain cylinder flute lying around somewhere it would be easy to knock up a spiked stopper for it. I haven't looked at the maths, but I imagine that wouldn't be an impediment? Now am I right in thinking that to give us an effectively bulging ("parabolic") bore, we'd need a thinned spike?
Sounds like we had the identical experience, Terry: better tuning at the cost of a seriously choked off voice. They still played (after a fashion), but they were not free-blowing and they felt like they were fighting me. Some players loved them, but I never could.
Yeah. I toyed with the idea of siting the wedge under the embouchure hole, so that the hole passes through the wedge. That would give one full revolution before the spiralling air column comes back around to the wedge. I wonder if that's enough to allow it to straighten out?
Andro
Posts: 191
Joined: Sun May 06, 2018 1:37 am
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Beginning flute and whistle player. Starting out seriously in Irish Traditional Music. Also interested in flute making.
Location: Melbourne, Australia

Re: PVC flute temperament

Post by Andro »

Hi paddler,

So are you saying that technically the flute bore is a conic section (truncated cone), and that the Boehm headjoint is a section of a parabola (which is, of course, a conic section, geometrically speaking)?

Apollonius of Perga clearly rules flute design!
User avatar
paddler
Posts: 752
Joined: Sat Nov 03, 2007 7:19 pm
antispam: No
Please enter the next number in sequence: 8
Location: Hood River, Oregon, USA
Contact:

Re: PVC flute temperament

Post by paddler »

Yes, I think they could both be correctly defined as conic sections. The Boehm head does not have a straight/linear taper, and in practice neither does the body section of most antique or modern conical bore flutes. When you profile these bores they have all kinds of deviations from a straight line.
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: PVC flute temperament

Post by Geoffrey Ellis »

Yeah. I toyed with the idea of siting the wedge under the embouchure hole, so that the hole passes through the wedge. That would give one full revolution before the spiralling air column comes back around to the wedge. I wonder if that's enough to allow it to straighten out?
This seems worth an experiment. I never attempted that variation (and in fact it never occurred to me!). One would have to factor the added chimney depth that would be created, which would not be uniform because of the taper in the wedge (though not grossly uneven I expect). But if you have to drill through the wedge for the embouchure hole, suddenly that chimney is going to be a couple of millimeters deeper, give or take.

I've never had the math skills to calculate the volume difference between a cylindrical headjoint and the parabolic version, so most of my wedge experiments were entirely trial and error. I made a wedge of a certain length and thickness and tried it out, checking the intonation and the voice. Modified it, tried again, etc. until I found something that worked. The tuning balance sort of tells you when you have it in the ballpark, but I do think that how and where it is placed has a big effect on the voice.

Now I'm sort of itching to try your idea, just to see what happens :-)
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: PVC flute temperament

Post by Terry McGee »

Geoffrey Ellis wrote:This seems worth an experiment. I never attempted that variation (and in fact it never occurred to me!). One would have to factor the added chimney depth that would be created, which would not be uniform because of the taper in the wedge (though not grossly uneven I expect). But if you have to drill through the wedge for the embouchure hole, suddenly that chimney is going to be a couple of millimeters deeper, give or take.
Indeed, and I thought that might be an advantage, given most of the suitable available tubing has pretty thin walls. And we don't want to make the outside diameter too great.
I've never had the math skills to calculate the volume difference between a cylindrical headjoint and the parabolic version, so most of my wedge experiments were entirely trial and error. I made a wedge of a certain length and thickness and tried it out, checking the intonation and the voice. Modified it, tried again, etc. until I found something that worked. The tuning balance sort of tells you when you have it in the ballpark, but I do think that how and where it is placed has a big effect on the voice.
I think trial and error is probably a very good approach. Firstly, we can't assume Boehm got it perfectly right (although it wouldn't surprise me!). But more than that, we don't know how people have dealt with his bore since. He presumably would have been aiming at a lower pitch, like A430. When his work was re-released by Rudall & Carte in England did they just take his bore, or did they retweak it for the high pitch coming into fashion there? And since then have all our manufacturers carefully reworked it for A440 or taken whatever they found as good enough? Given the work Albert Cooper and more recently Trevor Wye have been doing on improving the intonation on Boehm flutes, I would suspect that near enough has been good enough at many times. And of course we're using smaller holes than Boehm had in mind, so that's going to have some effects too. So the DIY approach is good!
Now I'm sort of itching to try your idea, just to see what happens :-)
Let us know what you find, either way!
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: PVC flute temperament

Post by Geoffrey Ellis »


Indeed, and I thought that might be an advantage, given most of the suitable available tubing has pretty thin walls. And we don't want to make the outside diameter too great.
Excellent point--I always added a lip plate to my earlier versions of this design, so that would render the plate unnecessary.
I think trial and error is probably a very good approach. Firstly, we can't assume Boehm got it perfectly right (although it wouldn't surprise me!). But more than that, we don't know how people have dealt with his bore since. He presumably would have been aiming at a lower pitch, like A430. When his work was re-released by Rudall & Carte in England did they just take his bore, or did they retweak it for the high pitch coming into fashion there? And since then have all our manufacturers carefully reworked it for A440 or taken whatever they found as good enough? Given the work Albert Cooper and more recently Trevor Wye have been doing on improving the intonation on Boehm flutes, I would suspect that near enough has been good enough at many times. And of course we're using smaller holes than Boehm had in mind, so that's going to have some effects too. So the DIY approach is good!
I found it interesting that in his book Boehm indicated that he actually preferred the flute to have a 20mm bore instead of 19mm. He compromised and did 19mm because it gave better intonation on the high notes (and it's possible that at least one important note was either unattainable or very, very flat with a 20mm bore), and this was what players wanted. But he thought the tone of the 20mm bore flutes to be nicer in the first couple of octaves.

As for the parabolic taper, I wouldn't be surprised if other makers discovered that it was good enough without much tweaking. I have reamers made to Boehm's specifications that I use for my headjoints (the very specs he provided in his book) and they work great. Not only do they work great, but they work great even when the diameter changes due to wood mobility. Boxwood is super prone to shrinking, for example. In fact, at one point I measured a bunch of headjoints (both metal and wood) from makers such as Cooper, Arista, Haynes, Powell, Altus, Brannen, Muramatsu and the amazing David Chu, whose wooden headjoints are considered to be among the best available. David's was amazing sounding (happened to be boxwood) and when I measured the bore it had clearly shrunk quite a bit, and yet it was superb. It differed quite a bit from the metal headjoints that had not shrunk at all (obviously), but I think it was nicer than most of them. I've seen the same thing with variations in my own headjoints, so I think that taper is pretty forgiving. As long as you start with an accurate reamer and reasonably stable material, even with mild distortion it still performs just fine.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: PVC flute temperament

Post by Terry McGee »

Yes, I've been intrigued by the notion of a conical flute based on a 20mm head bore ever since reading that back in the seventies! So I'm looking over Tunborough's shoulder with the thought that WIDesigner might prove the appropriate tool to model the rest of it.

And I think you are right - for some reason the Boehm bore doesn't seem very touchy. But then how did Boehm conclude that it should have the bulge we've been talking about? Again WIDesigner might enable us to replicate Boehm's physical experiments.
Tunborough
Posts: 1419
Joined: Sun Dec 05, 2010 2:59 pm
antispam: No
Please enter the next number in sequence: 10
Location: Southwestern Ontario

Re: PVC flute temperament

Post by Tunborough »

There's a precedent for this in the uilleann piping world: "rushing" a chanter to adjust the tuning. Has anyone ever considered using a helical coil (stainless steel wire or some other volume-filling material) in the headjoint to occupy the space, with a progressively tighter helix toward the stopper, and the helix coiled in the natural direction of airflow (clockwise down the tube)?
Tunborough
Posts: 1419
Joined: Sun Dec 05, 2010 2:59 pm
antispam: No
Please enter the next number in sequence: 10
Location: Southwestern Ontario

Re: PVC flute temperament

Post by Tunborough »

Well, I've got some results from modelling headjoints, and they are quite unexpected. I modelled a few flutes with cylindrical bodies, some 19 cm diameter, and some 20.5 cm diameter (3/4" PVC), with a few different hole geometries. I asked WIDesigner to lay out a headjoint around 120 mm long, giving bore diameters every 20 mm plus the stopper position, that would optimize tuning through two diatonic octaves, D4 through D6, plus cross-fingered C-nat for most trials.

If given enough latitude to choose the bore profile, WIDesigner suggests a headjoint that expands in the bottom 20 mm, contracts again to a bit less than the main bore diameter, expands again below the embouchure hole, then contracts again to the stopper. Often, there is no headspace at all; the stopper is right up against the embouchure hole.

If the headjoint diameter is constrained so it can't expand from bottom to top, WIDesigner suggests a 40 mm long taper around the middle of the headjoint, with a change in diameter of maybe 1 mm. Above and below this tapered section, the headjoint is cylindrical. With the right hole geometry, it is quite remarkable how much this simple tapered section can bring the predicted tuning of the octaves into line. I did try a less-than-optimal hole geometry for which WIDesigner also suggested a sharp taper around the embouchure hole, so the headspace was considerably smaller in diameter.

No sign at all of Boehm's parabolic headjoint. Is there any particular hole geometry you'd like me to try to tease out something parabolic?
User avatar
Geoffrey Ellis
Posts: 585
Joined: Thu Sep 13, 2012 1:15 pm
antispam: No
Please enter the next number in sequence: 8
Tell us something.: Crafting fine quality folk flutes from around the world since 1997, my goal is to create beautiful instruments that have the best possible voice, tuning and response by mixing modern methods with traditional designs.
Contact:

Re: PVC flute temperament

Post by Geoffrey Ellis »

Tunborough wrote:Well, I've got some results from modelling headjoints, and they are quite unexpected. I modelled a few flutes with cylindrical bodies, some 19 cm diameter, and some 20.5 cm diameter (3/4" PVC), with a few different hole geometries. I asked WIDesigner to lay out a headjoint around 120 mm long, giving bore diameters every 20 mm plus the stopper position, that would optimize tuning through two diatonic octaves, D4 through D6, plus cross-fingered C-nat for most trials.

If given enough latitude to choose the bore profile, WIDesigner suggests a headjoint that expands in the bottom 20 mm, contracts again to a bit less than the main bore diameter, expands again below the embouchure hole, then contracts again to the stopper. Often, there is no headspace at all; the stopper is right up against the embouchure hole.

If the headjoint diameter is constrained so it can't expand from bottom to top, WIDesigner suggests a 40 mm long taper around the middle of the headjoint, with a change in diameter of maybe 1 mm. Above and below this tapered section, the headjoint is cylindrical. With the right hole geometry, it is quite remarkable how much this simple tapered section can bring the predicted tuning of the octaves into line. I did try a less-than-optimal hole geometry for which WIDesigner also suggested a sharp taper around the embouchure hole, so the headspace was considerably smaller in diameter.

No sign at all of Boehm's parabolic headjoint. Is there any particular hole geometry you'd like me to try to tease out something parabolic?
This is very interesting. I'm not familiar with WIDesigner, but your findings remind me of the "optimized" Chinese xiao that was designed by Yan Lang at the University of British Columbia. I spoke of this elsewhere but he created an optimized bore that was a series of these expanded and contracted sections. The nature of the shape made it's creation impossible unless it were cast or molded somehow, since it could not be conventionally drilled or reamed out. I opted for casting.

A friend of mine is a math professor at a Vermont university (and an amateur flute maker) and he passed this data on to me. He is also experimenting with headjoints for the xiao and some of his theoretical models resemble what you've created. A sort of "undulating" bore profile that does not taper.

This seems to show that the "optimal" acoustic shape for some of these flutes might be quite different from what we are used to. But of course, it is easy to see why such bores are not really feasible with wooden instruments. Whether 19th century flute makers like Boehm were hep to these unusual bore designs would be interesting to know, but they obviously didn't choose to attempt them. Possibly the parabolic taper is the best (most optimal) design that can be created using conventional methods (drawn metal tubing or reamed wood)?

It would be cool to try to make the headjoint you describe and see how it performs.
User avatar
Terry McGee
Posts: 3330
Joined: Sun Dec 12, 2004 4:12 pm
Please enter the next number in sequence: 1
Location: Malua Bay, on the NSW Nature Coast
Contact:

Re: PVC flute temperament

Post by Terry McGee »

Woah, very interesting indeed. As I said, "we can't assume Boehm got it perfectly right". But equally, we can't assume WID got it right either, given it's pretty early days yet in its application to transverse flutes. But we owe it to everybody to find out, and, whatever we find, to progress it from there.

Tunborough, are you planning to do anything physical or should we detail somebody else to have a shot at it? Geoffrey? Anybody?

The kind of shapes Tunborough has mentioned are a bit out of our usual comfort zone, but I bet we could find ways to achieve or at least approach them, even if it requires inserting some separately turned and bored sections.

It's sobering to remember that if WID found that with the fiddling-the-head-bore approach, it might come up with some equally interesting suggestions for our older fiddling-the-body-bore approach.

I often have wondered how many of the older 19th century flutemakers gave the game away once the Boehm bore flute started rising into ascendancy. Maybe it's now time for oldies like me to be making my way quietly to an exit? Surely, at 71, I'm too old for change?

Nah, bring it on!
Post Reply