All whistles play out of tune

Whistle
1

A recent thread, /viewtopic.php?f=1&t=90803, prompts this observation.

Pretty well any note on any whistle will sound over a range of pitches, depending on breath pressure. A range of a semitone isn’t out of the question. So it’s possible to play out of tune on any whistle, if you want to work at it. On the other hand, it’s also possible to play in tune on lots of whistles. Sometimes you have to work at it. So what does it take so you don’t have to work at it?

I’m collaborating on some software intended for optimizing the tuning of wind instruments. I’m getting pretty good at predicting the limits for the range of pitches of each note on a whistle. But I’m not sure how to translate that into a whistle that’s easy to play in tune.

Here’s my current guess (expressed in terms of a D whistle):

  • Low D is in tune just a bit below the top end of its pitch range, before it jumps to middle D.
  • High D is in tune somewhat above the bottom end of its pitch range.
  • Between these two, the breath pressure to play each note in tune rises steadily and regularly as you go up the scale.

Does this sound like a reasonable objective?

2

Seems fine to me as long as you are talking about cylindrical bore instruments. Conical bore you might want to adjust there ideas a bit.

3

It would be interesting to test your assumption about gradually increasing pressure, by having good players who play in tune on various makes of whistles play up and down the scale and measuring the pressure in some way. I suppose in between the player’s mouth and the top of the whistle there would have to be some measuring device, I don’t know what.

It’s fairly common in Highland piping for people to rig up pressure gauges, and some reed makers test the pressure required for each reed, and sell the reeds as playing at such-and-such a pressure. So in Highland piping at least there’s an awareness of pressure and the value of measuring it.

Anyhow as a player I want a whistle where I can just play it and not have to think about compensating for an irregular scale.

There are a couple things I noticed over the last couple years, while owning/borrowing a large number of Low D whistles and testing them in various ways

  1. the approach to the tuning of the two registers varies considerably from maker to maker

  2. the range of pressure possible to sound notes in the 2nd register varies considerably from maker to maker

About the relative tuning of the 1st and 2nd octaves, some makers have the 2nd octave tuned relatively flat so that to play the octaves in tune you must somewhat underblow the 1st octave and strongly blow the 2nd octave.

Other makers have the 2nd octave relatively sharp so that to play the octaves in tune you must strongly blow the 1st octave and somewhat underblow the 2nd octave.

About the range of pressures possible in the 2nd octave, on the Overton Low Ds I’ve had you have a very wide range: you can blow the 2nd octave notes quite softly and they will stay up in the 2nd octave, and you can blow them very strongly and they won’t break into a higher harmonic. (But of course they’re only in tune with the 1st octave at one specific pressure, which is somewhat strong.)

On the Burke Low Ds I’ve owned, and on Burkes of other low pitches as well, notes in the 2nd octave have a fairly narrow range of possible pressure, which fortunately is right where they’re in tune with the 1st octave.

About it being possible to play in tune on any whistle, I have noticed that very good musicians with great “ears” will play quite in tune on pretty much anything they pick up.

I’ve done gigs with “legit” woodwind guys, guys who “double” playing sax, Boehm flute, clarinet, and various whistles and bamboo flutes etc etc. I’ve heard these guys play perfectly in tune on a given whistle, and when I’ve tried their whistle later I’ve discovered that the scale was pretty bad, off-the-shelf Generations and what not. These guys have “pefect pitch” and they hear the note they need in their head and they blow whatever they’re playing so that that exact note comes out.

One guy, a music professor at a university, told our class once that he got frustrated when he tried to make his own flutes out of PVC pipe. He couldn’t figure out where to drill the holes, because no matter where he drilled them he would play the flute in tune! (He needed somebody like me, utterly lacking “perfect pitch”, to try his flutes for him!)

4

I’ll chime in on this from a makers point of view. As Pancelticpiper has pointed out some instruments require more or less pressure between octaves to fit the scale and I have found this is due mainly to the size of the bore. Since most makers use off the shelf tubing we have to work with what is available and unfortunately the bore doesn’t always perfectly fit both octaves in the scale. Thus the need to overblow or underblow the second octave to make it fit the scale. Of course this situation is affected by the voicing but from my experience the size of the bore is the main culprit. To give an example, I was using a certain size bore for my low Ds and that particular size became unavailable as a result of my supplier reducing their inventory and it was not available anywhere in the US. So I was forced to use a tubing that had a bore that was just .008" smaller and the difference in the way the whistle played was amazing. I have had to accept the fact that the whistle is now a bit heavier as a result of the extra wall thickness but well worth it as the whistle plays much better and the octaves fit far better together.

There are ways to compensate for this imperfect bore size and many makers, such as myself, Michael Burke and others, are using bore perturbations to adjust the bore size at various places in the bore. This approach has long been used by Shakuhashi makers by narrowing the bore with lacquer at certain points in the bore.

Regarding measuring the pressure to play the scale one can use a system similar to a water level. I forget what this gizmo is called but it works like this… you attach a clear plastic tube to a ruler that is mounted on an angle and the other end of the tube goes in your mouth. The tube is partially filled with water (colored with some food dye so you can read it) and when you blow it registers a level on the ruler so that as you go up the scale the level gradually goes up. In this way one can measure the pressure needed to gradually ascend the scale and adjust the hole pattern to fit this pressure. However, I have found that there are other considerations when tuning a whistle such as particular notes which one would want to accent and lean into so minor adjustments in hole size and position need to be made.

So this is just my 2 cents and I hope it brings some clarification to the issue.

Ronaldo

5

Seems you are describing a manometer.

6

Yes, that’s the nub of my question. What does a “regular scale” look like on a whistle.

The manometer experiment, on a whistle of known geometry, might answer the question, provided the player wasn’t compensating at all for irregular tuning. For my purposes, we could dispense with the manometer and just measure the actual pitches the whistle produced when played the way the whistler wanted it to play in tune. With the whistle geometry and pitches, I can predict the air flow, at least in relative terms. That would be enough to test my hypothesis.

Oh, … interesting. How do you figure things would change? Dealing with bore tapers is definitely on the agenda.

7

Well, the makers out here can answer that better, but a simplistic answer is that the reverse conical bore evens out the differences in blowing needed to keep the whistle in tune across the entire range. For me it usually means I need to not blow hard in the high octave to avoid going sharp with the conical bore instruments (I play Copelands) but I do need to blow hard to keep the upper octave in tune when I play cylindrical bore instruments (how much is needed depends on the instrument).

8

Awhile back I was curious about the tuning of my 2 whistles. I think I only tested up to second register G and an a novice, but for what it is worth, I was satisfied when I put my Oak whistle against a tuner and blew the thing as sharp and flat as I could. After tuning it to the G as I play it, it seemed that within that range the “in tune” note hung out in about the same position within the extremes of the range from note to note. For me, playing an E.T. scale was close to playing in the center of the extremes. I could bend push and pull the tuning on the Dixon Trad much less, but had about the same results.

9

Conical bore you might want to adjust there ideas a bit.

Oh, … interesting. How do you figure things would change? Dealing with bore tapers is definitely on the agenda.

Well, the makers out here can answer that better, but a simplistic answer is that the reverse conical bore evens out the differences in blowing needed to keep the whistle in tune across the entire range.

While the physical playing trumps theory every time it is dangerous to assign playing characteristics to particular design features based purely on playing dissimilar whistles.

There is an equation (so this is theory) that was published over 80 years ago that predicts quite accurately the flat second octave of cylindrical whistles, the interesting thing is that this equation also applies to conical whistles and so I would expect that a conical whistle would also have a flat second octave.
So how do we reconcile this theoretical knowledge with the fairly common belief that conical is better than cylindrical for in tune second octave.
Well let me preface this with:
I have not yet made conical whistle
I have not yet designed a conical whistle and calculated the tuning
I have not had the opportunity to carefully measure a Copeland whistle

I do own a sweetone whistle but have not carefully measured or examined it - neither do I play it much
I do make cylindrical whistles that do not require blowing hard to achieve an in tune second octave

My belief is that many (most? all?) conical whistles are not pure ‘cones’ much like my cylindrical whistles are not quite pure cylinders, or as is well known Burke whistle are not pure cylinders, or modern Boehm flutes are not pure cylinders. A few perturbations in the bore, along with careful choice of bore diameter, head design and hole size can produce in tune instruments with either tapered or straight bores.

Now all that is needed is a definition of in tune…

10

What does a “regular scale” look like on a whistle.

For me it’s playing up and down the scale and the needle on the tuner doesn’t move… that’s what you need for “legit” gigs.

The manometer experiment, on a whistle of known geometry, might answer the question, provided the player wasn’t compensating at all for irregular tuning. For my purposes, we could dispense with the manometer and just measure the actual pitches the whistle produced when played the way the whistler wanted it to play in tune.

Ah, but the manometer would tell you WHEN the player was compensating for irregular tuning! It would tell you precisely how a good player with a good ear was able to produce an in-tune scale from a given whistle. No point in measuring the pitches produced, because with a good player those will be correct; in other words we already know what pitches a good player will produce.

11

I think your argument is circling back to my original hypothesis (which is good to hear). If the player had to use “irregular” pressure on some note, a lot different from the notes around it, then we’d conclude they were compensating. If this is true, then the player won’t need to compensate if we design a whistle to have a regular pressure pattern.

I believe you. The draft whistle optimizer tells me that we can make whistles that have a regular pressure pattern through two octaves, even through the C# and cross-fingered C-nat, except for two notes: the high C# will likely play flat unless compensated, and the fingered high C-nat (OXOOOO) will likely be too flat even with compensation. I haven’t made enough examples to test this out (spending more time building software than whistles), and even if I did, I’m not a good enough player to serve as a “reference player”.

I haven’t tried the optimizer on a conical bore whistle yet.

12

and the fingered high C-nat (OXOOOO) will likely be too flat even with compensation.

I’m not sure I own a whistle that 2nd octave cross fingered C-nat works on - I use half holed C-nat for the 2nd octave and often for the 1st octave as well.

13

This is all entirely dependent on design of each whistle, and the execution of those designs, respectively. One thing to remember about this is that different makers have different ideas on tuning, hence the many subtle differences that can be found in say, narrow bore brass D’s based on the same bore size. The actual playing characteristics of one whistle may be favored by some and liked less by others. Bore size also adds another element, because it often changes everything in terms of breath requirement and voicing…

I am not a machine, and neither are you. Your idea of what constitutes a steady rise in force of breath and mine aren’t necessarily going to be the same when it comes to actually playing. The bottom line is if you sound in tune, especially with other instruments around you, then you are… and at that point the whistle is a non-issue altogether. Which brings me to…

A perfect whistle is only perfect in the hands of a perfect player, playing perfectly. I’m going to go play my Feadog before I go to bed. It’s good enough for me. :wink: