OK, so I've been experimenting a bit with the Kwela style of blowing and I can say that it definitely has the potential to produce a much louder, richer and edgier
bottom octave on most whistles. I have experimented with the following whistles, all high D:
Sindt
OZ vambrace
Generation
Clarke Original
Clarke Sweetone
Shaw
Feadog first generation
Feadog second generation
Freeman bluebird
Freeman blackbird
Oak
Guinness
Walton's Mellow D
The effect can be obtained on all of them, but it requires some nuanced adjustment of whistle angle, depth of insertion and lip shape. Each whistle has its own
response and some are much more difficult than others.
It seems to be important to have the whistle at around a 45 degree angle, which makes me wonder if a certain amount of asymmetry in the shading is a factor.
When holding the whistles at just the right angle I was able to increase the volume of the lowest notes on some whistles by a huge amount -- almost to the point of
blowing as hard as I could, at which point it was really loud and still not breaking to the next octave. A warble starts to occur on some of them some of the time.
The Clarke Original was not as easy for me to get the effect as most of the others, but I think this was because I had bent and tweaked the blade to the point where
it was almost perfect playing normally. I think I see why Jim Donoghue spent a lot of time messing with his Clarke whistle. You can change their behavior quite
dramatically with very minor adjustments, and I suspect the adjustments you need to optimize playing shaded are different to those for normal playing.
None of the whistle are very comfortable to play this way (with the head inserted far into the mouth), but I think you probably do need to do a certain
amount of dynamic embouchure adjustment to keep the balance just right across the full range of notes. Whether this is something that would be acceptable
to the average whistle player is a separate question. I suspect probably not.
So getting to stringbed's politely ignored question, which I hadn't actually ignored, but was pondering about because it seemed to be right on point,
I think the approach that will be more likely to optimize the end result would be one in which the head is designed in such a way as to make this style of
dynamic shading easier. However, for widest community acceptance I suspect that trying to take some insight from the physics/acoustics that is going on here
and trying to build some of that into the hardware would be the way to go. In either case, it is probably going to mess with the tuning, because a significant amount
of flattening seems to occur, especially if you don't compensate with a lot more pressure.
Are you planning to start making whistles with a flat foot Terry?