UNDERCUT Whistle Blade - seen any good examples?

Whistle
1

Ever seen a whistle with an undercut blade?
I’m very curious about the possibility or usefulness of an undercut being put on a whistle blade (the sharp edge in the window where the air is split after it leaves the mouthpiece’s windway). Almost all whistle blade designs are overcut designs; are angled only on the top side, so that the lower side of the blade, facing the inside of the whistle, is the unaltered inner surface of the tube itself, be it rounded or flattened.

This curiosity comes about from being a life-long flute player. A flute player can vary the direction of the air flow, from mostly downward into the tube, or much higher across the top of the tone hole. Very different tonal effects in both octaves can happen when you direct the air stream in different ways, either up/down or left/right (off of direct). A whistle with an undercut might allow higher air stream pressure to be directed downwards, and this may alter the tonal balance or allow the whistle to increase it’s range of possible tone and loudness under different air stream pressures or staccato note attack pressure. I’d love to hear if this has been experimented with or if there are examples of undercut blades on high whistles or low whistles!

A whistle mouthpiece typically prevents the player from adjusting the angle of air stream approach to the blade edge, an undercut would provide some kind of new option. (The Japanese shakuhachi mouthpiece design is the only design I know that provides the typical whistle edge while allowing the player to adjust the airstream approach to the blade.) The undercut/overcut design balance could be adjusted to make sure the resulting playability in the two octaves achieved a good balance. Also, if the undercut angle was held below a certain level, hopefully any slight increase in chiff, from having a surface for the airstream to flow past, would be kept very minimal. If the edge of the blade was metal, it could be so sharp that any chiff added by an undercut would be balanced out by having chiff reduced in another area; from the edge of the blade.

I searched “Undercut” and found lots of use of the word in the Flute Forum, mostly about key holes, nothing about whistle blades.

Thanks for any thoughts on this at all! It’s a very rare issue. Cheers!

2

Are you describing a Native American flute? Note the splitting edge.
https://upload.wikimedia.org/wikipedia/commons/thumb/c/c8/Native_American_flute_Composite_Anatomy.jpg/400px-Native_American_flute_Composite_Anatomy.jpg

3

As Tommy says, undercutting is a characteristic of NAFs. It may be related to another characteristic … switching between registers can be very difficult on some NAFs. In my experience, if a whistle blade is higher up in the airstream, it is harder to hit the second octave; undercutting the blade may lead to the same effect. This isn’t a problem on NAFs, but is undesirable in a whistle.

At the other extreme is a whistle with no undercutting or overcutting: the airstream hits a 2 mm high wall at the far end of the window. I have a whistle like that on my desk, and it plays just fine. The break between registers is quite clean, with only a slight tendency to wobble in the transition.

4

Undercutting is really depending on how the rest of the whistle is made.
Lot of factors involved (here’s a few - but not limited to this)

  1. Shape of the windway.
  2. Type of windway chamfer.
  3. Is the windway floor at the same level as the inside bore surface?

As the other posters stated, it is much more common on Native American Flutes.
This is because Native flutes are usually made by cutting a duct that makes the windway on the surface of the flute. The floor of the windway can be much higher than the bore surface. It’s called a “raised nest” and it usually requires undercutting or an “underramp”.

5

Thanks for those responses. I’ll check out more of the First Nations flute styles. How they handle that, I haven’t seen in particular, but it sounds like the undercut is about 100% of the blade.

I am still curious though, about using a partial undercut on the Irish/folk style whistles, be they wood, metal or PVC. The undercut would likely be blended with the normal overcut, in some ratio that could produce peak volume and tone, with regularly. So maybe it would be 75% overcut, with 25% undercut? I’m just wondering if any whistle maker has done R&D on this and includes a blend of over/undercut on current models.

My suspicion is that some undercut would help make the production of tone more efficient, by forcing the start of the air vibration in the tube with more vigor. Less % of the air would be escaping above the whistle. I presume that air going up above the whistle blade is essentially wasted because it’s not making music. When pressurized air in the windway reaches the window, it will depressurize and expand in all directions available (up, down, left, right), so how the whistle minimizes loss of air and music-making potential there, is important. The edge of the windway should be very close to the blade, to assure the air is hitting the blade with maximum efficiency at converting air flow to musical tone.

I would love to see a video made of air flowing through a whistle, perhaps if a whistle player took an inhale of smoke we watched the air stream in slow motion hit the blade and then how much goes into the tube, vs fly up above the blade. Another weekend science experiment for someone who’s had a few pints. LOL

6

I presume that air going up above the whistle blade is essentially wasted because it’s not making music.

In short. No. All the blade does is make the air-column vibrate by splitting the air stream, so the air going above the blade is not “wasted”–it is in fact essential to the sound production. The best results are achieved if the height between chamfer-floor and underside of the blade is very small. A larger height means, you need more air. I have experimented with an undercut on my whistles but the results were less than ideal. I think every tweak or “trick” that can be done on as simple a construction as a whistle, can and has been done by now.

The edge of the windway should be very close to the blade, to assure the air is hitting the blade with maximum efficiency at converting air flow to musical tone.

No. If the edge of the windway is too close to the blade the whistle will flip too easily to the 2nd octave or not sound at all. It will also have very little backpressure so control of the whistle will be less than ideal. You need a certain amount of backpressure to make the whistle controllable. There is a “sweet spot” in distance between windway edge and blade. For a high D it is somewhere around 5mm. Give or take a millimeter. A very small change can make a ton of difference in tone and playing characteristics. I made whistles with longer windows, which makes them more airy and chiffier sounding. They also need more air-speed for the 2nd octave.
A lot of that stuff is explained very nicely here: https://sites.google.com/site/guidogonzato/ggwhistles
Useful info is also contained in the Wikipedia entry “fipple”: https://en.wikipedia.org/wiki/Fipple

7

… or is working toward a PhD …

Patricio de la Cuadra. The sound of oscillating air jets: Physics, modeling and simulation in flute-like instruments. PhD Thesis, Stanford University, 2006.

No videos in the thesis, but he does have pictures, using Schlieren flow visualization.

8

Next up: Filling violin bodies with lead to improve resonance and square wheels on cars for improving traction!

Reviews from Perlman and Hamilton to follow.

9

The physics isn’t obscure (it’s why aeroplane wings work to give lift for instance), though I only really remember my school physics.

It’s the Bernoulli principle, which gives an explanation of how a stream of liquid or air moving moving from say a larger tube to a smaller one, or a wide river to a narrower one, speeds up without more energy being put into the system. The faster moving air/liquid is at a lower pressure. It’s the principle describing how the windway in a whistle creates fast flowing air, and why there is turbulence created by the labium.

At the labium, the air flow (jet stream) is split in such a way that one side has to travel further, therefore lowering the pressure on one side of the labium (same as how an aeroplane wing creates lift with air moving faster over the top of the wing). Because there is an imbalance of pressure (in a partially closed system), the system tries to equalise the pressures but never succeeds in a tin whistle because it is set up so that the equalisation is continually overshooting each way. This oscillation creates a standing wave inside the tube of the whistle. The length of the tube gives the frequency of the standing wave.

As Sedi pointed out, the air stream split to the outside of the labium is not wasted, is is utterly essential in the Bernoulli effect. Also the air coming out of the windway does not immediately dissipate in every direction as you were surmising, it takes the form of a jet stream playing onto the labium.

OK, now someone correct me or explain it better. I’ve probably remembered something wrongly … :party:

10

I don’t know quite what you mean by “typical whistle edge” but endblown flutes are very common throughout the world, the shakuhachi is only one of hundreds of species.

There are all the species of notched endblown flutes like the shakuhachi and the other Asia species and the Kena in South America.

Then there are numerous species of endblown flutes lacking a notch, where the edge is cut all around the top of the flute, like the kaval in Bulgaria and the ney (in various spellings) in the Arabic world, Persia, Turkey, etc. Oftentimes the edge is rounded, while especially on some modern instruments it’s a flat angled edge such as is used on Irish whistles.

Here’s a great demo of the Kena/Quena well showing the notch and overall construction. There’s a node in the cane at the bottom that they drill a hole through, the Kena therefore has a slightly conical bore. Also there’s a thumb-hole

https://www.youtube.com/watch?v=aQhqop2Ej84

The Ney- as you see there’s no notch, the top of the tube is angled all around, so you can rotate the headjoint to any position without affecting how the flute plays.

https://www.youtube.com/watch?v=Dv_yow9J6gY

Kaval, note the upper-hand posture is identical to that of the uilleann pipes, and the “pipers grip” used on Irish Low Whistle, while the lower hand position is different from the Low Whistle “pipers grip” due to their being four holes instead of three

https://www.youtube.com/watch?v=ESYRozEP1s8

As far as I know all of these endblown flutes have the angle/slope of the cutting edge on the outside of the tube, while the inside of the tube is an undisturbed cylinder as is seen on Irish whistles.

About Native American flutes, as far as I know the pre-contact flutes were made and played like the Ney and Kaval, open ended and lacking the complex setup that modern NAFs have. Some tribes like the Miwok still make flutes in the old style.

Here is a photo clearly showing the identical embouchure, and rounded top edge, as used on kaval

11

Here’s my 2 cents about how a fipple creates a tone. First off I hear a lot of comments about the “splitting” of the airstream which to my way of understanding doesn’t actually happen. From my point of view as the airstream flows across the window it creates a low pressure in the bore and the airstream moves into the bore, then as a result of resistance (drag) in the bore the pressure increases and the airstream moves outside to ambient air which is lower. This continues at a fast rate and the result is oscillation of the airstream and not splitting. I’ve heard the term “air reed” in relation to how a fipple creates a tone which seems to me a very appropriate name as to how this works.

Some time ago I saw a video demonstration of this in a wind tunnel using smoke which was slowed way down and one could actually see the oscillation taking place. And there you have it an “air reed” oscillating over an edge creating a tone, which as Pancelticpiper has referred to works on most all end blown flutes.

And by the way the bit about undercutting the labium simply adjusts the point where the oscillation happens and directs the airstream downward into the bore. This is a technique that I use to ease the resistance in the oscillation. If one looks closely at the Overton/Goldie design or the Copeland design one can see that the ramp is bent downward to direct the airstream into the bore.

Ronaldo
http://www.reyburnwhistles.com

12

The fact that the air stream (jet stream) is being split can be demonstrated both anecdotally and by experimentation.

Blowing across the top of a bottle, we know that we must get the angle right, with the right amount of air crossing the bottle top and intersecting with the opposite edge to get it to resonate. We also know that blowing a transverse flute to get it to resonate means directing the air stream at the edge on the opposite side of the hole to the player. I have a Quena, when I play it I know that I have to get the correct stream of air onto the labium, again, intersecting with the labium.

If what you propose were true, then simply taking the fipple off a Generation whistle and blowing down the tube would create a whistle sound because of the holes in the tube. We know that does not happen.

The explanation truly is the Bernoulli effect. Like an æroplane wing where, because of its shape, the air has to move further and therefore faster across the topside of the wing creating lift, the whistle’s labium splits the air stream and creates a pressure imbalance. There is an oscillating siphon effect (because this is not a wing) were the system tries to correct the pressure imbalance.

If the “drag model” were true, then æroplanes would not fly. The Bernoulli effect has been known for over 200 years and really is not controversial.

13

For those interested, perhaps this will help clarify:

http://www.flute-a-bec.com/acoustiquegb.html

14

Thanks, Loren.

Having spent a career around pipe organs, most stopped wooden pipes are made with the mouth in the usual way, but there are open wood pipes that are made with inverted mouths for whatever reason. These are often called ‘Melodia’. There’s also a triangular variety that uses an inverted mouth. Anyone make a triangular whistle?

15

Thank you Loren for the link which clearly shows that the airtream does not split but oscillates over the labium which is precisely my point.

16

Thanks Loren, that is an interesting article - I will need to digest it in full this evening.

Quick question – is it implying that the returning pressure wave in the tube is (a part of) the cause of the switch of direction of the airstream?

17

I have a short answer and a longer answer, but the shorter is all I have time for before work, so here it is:

I think that yes, it would seem what you suggest is being implied, however I’m not entirely certain. If I have the time/energy later today I will try to elaborate. That said, I definitely don’t have all the answers due to my lack of degrees in fluid dynamics, particle physics, and acoustical engineering :sniffle:

18

Based on my reading of the research, the answer is an unqualified, “Yes.”

A note about the source… It seems to me, Philippe Bolton has contributed enough to PhD theses to earn one for himself. In particular, he sat on the jury for the 2009 thesis of François Blanc.

19

Thanks both forconfirming the impression I had reading this over breakfast. I have been trying to sort this out in my head, and get the physics straight. I am aware that any explanation is an approximation butstill have a few niggles in my head.

First, I am completely comfortable with the idea that the vibrating air column in the recorder can influence the oscillation of the air across the labium in a way that reinforces the air column vibration. A returning pressure wave will could easily push the blown airflow from the inside to the outside of the labium. It could also be argued that it is the change in airflow across the labium that drives the vibrating air column, but that makes it more difficult for me to see how the frequency is controlled by the open holes. My guess is that the two effects are locked into a coherent whole where neither entirely causes the other.

What I am trying to to figure out how this oscilation starts in the first place. I cannot see the air column in the recorder playing much part - so it must be the direction-switching nature of the airflow across the labium. What causes this? It is a bit like the starter motor on an infernal combustion engine - needed to get things going but irrelevant afterward?

Or am I lookig to deeply?

20

Second question first: the air stream from the windway is inherently unstable. Any little wobble–and there will always be wobbles–will grow as the stream passes across the windway. The dominant wobbles tend to be up and down (in and out of the window). Section 2.2.1 of Patricio de la Cuadra’s thesis describes this, and points all the way back to Lord Rayleigh.

The wobble at the window will propagate down the bore. The bore resonances, which the open tone holes control, will reinforce wobbles that have certain frequencies, until the whole system is oscillating at one of those frequencies.