In the course of modelling whistle designs, I’ve come across an issue with the (somewhat) conventional fingering OXOOOO for the second-octave C-natural. With some patterns, OXOOOO does not give a solid note, or sounds much higher than intended, while OXOOOX does give a reliable note.
Looking at impedance plots, I can see what’s happening: there are actually two resonances involved. When OXOOOO is a problem, the frequency of the first resonance is less than the second harmonic of the lower register. The frequency of the second resonance is quite a bit higher than the second harmonic. This suggests that when the whistle is blown from the first register into the second, it is more likely to jump to the higher resonance than the lower resonance because it’s already got a lot of energy at the second harmonic. OXOOOX lowers the frequency of both resonances, makes the first resonance smaller, and brings the second resonance closer to the second harmonic, so the whistle will jump more readily into the second resonance. For example, for a particular whistle (a whistle in A, under construction and therefore still flat), my model suggests the whistle could play in the following ranges (depending on how hard you blow):
OXOOOO: 665 to 756 Hz, 1363 to 1426 Hz, 1644 to 1668 Hz.
OXOOOX: 665 to 756 Hz, 1309 to 1329 Hz, 1453 to 1510 Hz.
For both of these fingerings, the second harmonic of the first register would be around 1500.
For OXOOOO, the real whistle wouldn’t play around 1400 Hz, only around 1650 Hz. For OXOOOX, it would play from 1460 to 1525 Hz in the second register.
As I drill the holes larger to bring the whistle up to the design pitch, the problem is getting better. But I’m curious how to approach this phenomenon in general …
Never play high C-natural: it annoys the neighbours and makes your eyeballs bleed.
Forget about OXOOOO; always use OXOOOX (or some other fingering convention).
It always happens before the whistle is finished; it will go away once all the holes are the right size.
It won’t happen if you follow this rule of whistle design: …
Well if you found a good working x-fingering for the high cnat, where’s the problem? Don’t know if there’s something like conventional fingerings up there at all… there’s not even one definite fingering for the middle cnat. FWIW, depending on the whistle I’m using oxoxxx or oxoxxo. oxoooo never gave me good results; it may be (more or less) in tune but usually sounds terrible.
Generally I use either OXOXXX or a C-nat thumb hole for high C nat.
I thought OXOXXX that was the conventional fingering for it.
OXOOOO and OXOOOX don’t work reliably, if at all, in my experience.
What does your software say about OXOXXX fingering?
My experience is similar to Colomon. The half-holed C nat always works. I want use either the half-holed DOO OOO (preferred) or the OXO XXX for the cross fingered C nat up there - context sometimes matters for the choice between the two. Stanton’s other fingerings should be close as well. On a lot of whistles there is little difference between the OXO XXX and the OXO XXO. You seem to need the BH1 and BH2 covered to get the C to speak reliably IME. So I really never tried to make a whistle that could use a OXO OOO or OXO OOX fingering. Don’t think I’ve used the OXO XOX fingering either - could come in handy. Thanks for that.
Point taken, but the challenge I’ve taken on is a reliable cross-fingered C-nat that is good enough to keep you and pancelticpiper happy.
In short, it says .
For the specific whistle I’m making, it should produce a G-natural that is bang on target. In the general case, the second register resonance looks to be at or close to the second harmonic of the first register, even under construction, before the holes are drilled out to full size, so it should always speak reliably.
Tunborough, I am still a bit curious. I’ve not known the OXO OOO fingering to be conventional, even somewhat, on the whistle. For recorders in C, yes, for the C above the bell note having the pinch hole closed. Which is more akin to a first octave C on a D whistle. But I’ve not seen that fingering suggested on a whistle for C nat at the top of the second octave. I’m curious if that is a fingering you are familiar with on a whistle. Or was that fingering a design goal?
Fair enough, I guess…but am I really that obsessive? [looks in mirror]
Also, Tunborough, this sounds like a very interesting program you’re using. I’ve been interested in software that would calculate the effect of various fingerings on my whistle designs for some time. Is this a program that you wrote yourself? Is it available Out There somewhere? Could you give us a link?
I confess I don’t remember where I got that fingering from. I needed to pick a fingering; maybe I got it from a teacher, or maybe it just seemed like a plausible choice at the time. Going forward, I’ll work with OXOXXX.
The software is a collaboration with two other people. Our intention is to let it loose on the world at some point, but I don’t think it’s ready for prime time yet: it isn’t packaged as a standalone program, there’s no user documentation, and the user interface leaves much to be desired. (It’s more fun to play with it than work on user interfaces.) And although I’ve been working on whistle modelling since I joined C&F, it’s only been in the last week that I’ve trusted the software enough to drill holes based on its recommendation.
I will, of course, keep you posted on developments.
Don’t know if there’s something like conventional fingerings up there at all… there’s not even one definite fingering for the middle cnat. FWIW, depending on the whistle I’m using oxoxxx or oxoxxo. oxoooo never gave me good results; it may be (more or less) in tune but usually sounds terrible.
