Tweaking a whistle with a dirty/harsh sound

Whistle
1

I have seemingly had very good luck with Generation whistles, because despite the many complaints about them off the shelf, the half-dozen or so that I have in various keys play quite well. Or I should say, most of them do. I have one, a brass Eb, that has what I can only describe as a “dirty” sound. Here’s a quick tootle, hopefully you can hear what I mean:

https://voca.ro/1lsbSuOkqw7t

I like a bit of chiff, but this almost sounds like it’s got a frog in its proverbial throat or something. The impurity in the sound in this particular whistle has a harsh quality that I’d love to smooth out a bit if I can. The other Generations I have have a much more pleasant chiff to them, like a breathy vocal vs the rough gravel of this one.

I’ve decided to use it as a bit of an experiment in tweaking, only I’m getting a little discouraged. The first thing I did was to check for errant bits of plastic, which I figured might be the cause of the tone. Nope, all clear! Then, I filled the cavity as some people suggest, but either I didn’t do it right or that has little to do with the purity/dirtiness of tone, because it’s still there.

Is there something specific I should be looking for to clean up that sound?

2

Check the wind way exit to see if is the same size height at both sides. Sometimes Generations or Walton’s have a little more height on a side and it makes the air stream exit uneaven. Then the blade receives more air on one side than the other. The wind way exit needs to be a near perfect rectangle.

3

Sometimes there’s flash inside that can be trimmed off with a thin Exacto.

Sometimes the various parts of the mould didn’t align properly.

I’ve done a fix on a couple Generations that works very well to improve tone, efficiency, and tuning:

  1. Saw the head in two, right through the window.

  2. If the 2nd octave was flat you can file off more material (some is lost in the sawing) on one or both of the exposed areas. Shortening the window raises the 2nd octave.

  3. Glue the head back together. The best glue is the old-school styrene cement (which I think is illegal now). You have a brief period of time when the glue is set enough that the whistle can be played, but the alignment of the two halves of the head can be adjusted. I align them to get that pure clean birdlike 2nd octave, and a round full low octave.

The blue top is off-the-shelf and plays fairly well, but has a flat 2nd octave.

The red top has been sawn through and glued back together. It’s a superb player with strong low notes, sweet high notes, a pure clean tone, and the octaves in tune.

Yes it’s not much to look at! But my instruments aren’t for display, they’re tools to do a job. I don’t care what my hammer looks like either.

(Note the beginning of a split on the socket of the blue top. That’s why I bind heavy black thread around the socket, then soak it with super glue, making it impossible to split.)

4

Jerry Freeman will glue a small square of material on the underside of the blade on a Generation head. It serves two purposes. It shortens the window by bringing the leading edge of the blade closer to the windway, similar to Richard’s modification. It also lines up the bottom of the blade closer to the floor of the windway, which can also help clean up the tone.

5

I was up in Canberra on business in the early, scary Covid days, when I suddenly found myself with some hours to fill and nothing to do. This was not a time for looking up old friends! So I went to a big music shop and asked if they had any tin whistles. And they offered me a Feadog. I didn’t have to ask if I could try it out first, indeed they made it very clear I wasn’t to play it inside! So with not much choice, I bought it and headed down to the nearby park, found myself a park bench, unpacked my new whistle and blew it.

It was bloody awful! I could get all the notes, but all of them sounded badly. Like bigsciota’s, but worse. I toyed with the idea of bringing it back, but couldn’t really see much point. What would a bunch of electric guitarists be able to do with it? They would probably blame the player! And it would still leave me with nothing to do for the next few hours. So, I decided to man it out, work as best I could within its appalling limitations, and consider myself lucky that better whistles awaited me back home. If only I had thought to bring one!

As I persevered, I learned to avoid the worst of the note and pressure variations that gave the worse results, and I started sounding a bit musical. But the thought crossed my mind - am I learning bad habits that I won’t be able to shake later? Fortunately it seems not. And I was encouraged when people out walking their dogs or wheeling their babies came and sat on nearby benches, and thanked me for brightening up those dismal days.

Anyway, when I got back home to the coast, the Feadog was consigned to the Cupboard of Doom where it has sat till yesterday, when I dug it out and examined it closely. Yes, it was one of those heads that when you look down the windway towards the sky, you see a lot of sky and not that much ramp. The perfect example to test Tunborough’s suggested Jerry Freeman tweak.

I thought that a piece of plastic credit card would be about the thickness I needed, and would offer water resistance and ease of trimming. I measured the head and found I needed a piece of material about 8mm wide and 6mm long. Interesting challenge - how to position and glue something that small and light inside somewhere so small and dark. So, I decided to leave it on the end of a longer piece, use that to position it, and then cut it off later. I cut a strip off the end of the card the required width, marked 6mm in from one end, and offered it down the windway, guiding it under the ramp. Perfect! But now how to keep it there and press it up under the ramp? I found a scrap of plastic rod that was just the right diameter when poked under the card. If that hadn’t been right, I could have whittled a stick or jammed in some rubber or something. Ah, just thought of another solution - a pair of long nosed pliers holding the card in place, with a rubber band around its handles to keep them closed while the glue cures. Or better still, a pair of surgical forceps, with the rubber band wrapped 3 times around the handle end, but slid down towards the hinge so it does nothing at first. Locate the card scrap, slide the rubber band up and let go. Hmmm, you could possibly cut out just the final piece of card you need (with some spare sticking out into the window), stick it on one jaw of the forceps with poster putty, put a spot of glue on it, carefully guide to place and lift onto bottom of ramp, slide the rubber band up and let go. Might try that next time!

I pulled the card back so that the 6mm mark was back at the exit of the windway, and put a drop of Tarzan’s Grip Supaglue on the visible section. Then pressed it down with a toothpick, and guided it back under the ramp until I reached the 6mm mark. Then pushed the bit of plastic rod in to jam the card up under the ramp. And set it aside to cure.

This morning, I had to face the question, how to cut off all the leftover strip of card filling the window and sticking out of the blowing end? I opted to drill a series of tiny holes through the card across the middle of the window, being careful not to reach the sides. Then poked my old standby, the No 11 Scalpel blade, through the holes in the card, severing the leftover bit from the bit glued under the ramp. The leftover bit pulled out easily. Thinking about it later, it would be possible to drill those holes before gluing the card in place.

Just for fun, I tried the head back on the body. Crumbs, it was already better than it had been, though very quiet!

So then the fine work. Using the scalpel, trim the leading edge of the card back to open the window to about 5.5mm in length. The plastic card works well under the scalpel. Then “sharpen” the leading edge a bit. And testing the whistle now, it plays perfectly up and down the range. Looks a bit funny with a bit of card visible at the back of the window, but who cares.

And here’s the interesting bit. I compared it to a recent Walton’s Irish Whistle, an older Feadog and some Generations of various ages, and it’s now the best of them. And looking up their windways towards the sky, I see a lot of sky! And looking up my doctored Feadog, I see the barest glimmer of sky.

Now, my method outlined above is not the only way of doing it, and may not be the best. It would be interesting to hear how other people, Jerry Freeman included, locate and glue the card or whatever they use for lowering the bottom of the ramp.

And looking at these other whistles I have lying around unloved, I won’t be throwing out the rest of the credit card!

6

I think I had a variant of that problem in a reasonably old Generation that caught my eye in the Cupboard of Doom collection. But it had other issues too, and was really quite unplayable. So I decided to have a play with it.

It does suffer from probably a bit too much sky when looking down the windway towards the light, but nowhere near as much as the Feadog above. So I’d need to find something thinner than the credit card. But I figured that wasn’t the real problem.

And you got the impression that the bottom of the ramp was not really parallel with the bottom of the windway. Though not dramatically off.

More concerning I think was the very narrow windway exit - only about 0.7mm high. Especially as some of that is sky. So we’re not including much ramp.

And the top of the windway seeming to dip down a bit in the middle.

The bottom of the ramp was a bit hacked too, perhaps from a previous owner having tried to make some changes. Or maybe even me?

Of interest is the trouble I had getting it off the tube, and then the trouble I then had opening up the socket to accept the tube again. This is one very shrunken head - perhaps the worst I’ve come across. And so it’s possible that we’re seeing some asymmetrical effects of all that shrinkage. A fair assessment would be that it’s a write-off, but that only challenges me to give it a shot.

So, I opened up the windway exit, concentrating on flattening the dip in the ceiling, reducing the tilt, and raising the ceiling, rather than lowering the floor (we don’t want more sky!) I won’t pretend I fully succeeded, but it’s now playable, whereas before it was complete rubbish. More attention to detail might get it sounding better, including perhaps some sky reduction, but it’s probably really not worth further effort. Interestingly it seems to play better with some tubes than others. I don’t know what that’s about! But I guess it’s telling us that it’s still a bit marginal.

7

I found an old Gen Bb that has the problem of the mucky sound. I think it’s caused by the issue Tommy describes, of an uneven, wispy-edged ramp.

Some of my family were cutlers, so I recognised the name of an excellent manufacturer of jewellers’ needle files when I spotted a bundle in a junk shop, barely used and priced at £3. As soon as I discover where I put them when I moved house, I’ll have a go at gently levelling it off. The whistle has the best octave balance of all of my small collection, and if I could get the tone a bit cleaner, it’d be a gem.

The apparent crack in the head doesn’t appear to penetrate the plastic far or even extend all the way to the edges, so I think it’s just a scratch:

https://www.dropbox.com/scl/fi/qdjjenpcvwmhq5oczed2g/Gen-whistle-head.jpg?rlkey=yuei2k8pdln51t8x6mswdy5e5&dl=0

But why, after decades of making them, has the manufacturer not worked out that this makes their whistles sound like a cork stuck under a door?

Maybe I answered my own question in another post. They’re not instruments, they’re toys, and as long as people keep buying enough of them there’s little incentive to invest in improvements.

It wouldn’t do for me, though.

8

Maybe I answered my own question in another post. They’re not instruments, they’re toys, and as long as people keep buying enough of them there’s little incentive to invest in improvements.

You are making the mistake of thinking that if some people appear to regard these as toys, they must be. They are not and are not made to be toys. To deny this is to ignore the numerous musicians who play these and the wonderful music they play.
Generation and Feadóg heads have been redesigned several times over the years. We can argue whether the redesigns improved them, but it is fair to assume the intention was to make them better in one way or another.
Injection moulding is an imprecise with not the most consistent outcomes, bear that in mind. It’s at the root of the problem.

9

Hmmm, I wonder if you’re both right. They are musical instruments, but on a toy’s budget. I wonder if doubling the cost of the instruments would enable them to be made significantly better. But would it also halve the sales?

We haven’t even been able to decide who buys the things (apart from us in the Irish music community) so we are not in a great position to posit ways forward. Presumably the companies have arrived at a quality/price tradeoff that satisfies their needs, even though it clearly doesn’t satisfy ours.

Do we need to form an International Whistle Survivors Mental Health Support Group to lobby the Pope to intervene? Or send a delegation to Brussels to demand Minimum Whistle Standards. Oh God, I can hear Nigel Farage already…

10

This is partly what puzzles me.

I have two Feadog Ds that I got decades ago. Now I’ve seen photos posted here, I know both are the Mark I shape. One played okay, the other sounded rough; a modern tuner tells me they were both sharp, but the one I didn’t like much was really sharp. Pushing the heads further back has made them into nice whistles that even a rank beginner like me can play in tune. I use them all the time for practice because they’re as light as a feather, not especially loud, and they sound lovely.

The modern Feadog I bought recently as a car whistle arrived pitched more accurately than either of the old ones. The plastic of the head appears harder and glossier, and the edges and seams are noticeably cleaner. The metal body might be a bit lighter, but there’s not much in it. But it sounds 'orrible, and I can’t seem to improve much through learning its quirks. Obviously a decent player could do a lot better, but even I can make the old ones sound all right. None of them has been tweaked except for moving the heads so they play the scale in tune.

So with more advanced plastics and technology, you somehow seem to get worse sound.

I suppose one question would be, has the relationship between the sale price and the cost of production changed? Did they effectively used to be more expensive whistles because, by comparison, more was invested in making them?

To some extent companies can take their eye off standards once they’ve established their place in the market (which Feadog were presumably still doing when they made my old whistles), but it’s risky do it for too long unless they want to be on a downward slope. But I haven’t tried their Pro model, and maybe that’s where they’re focusing attempts to improve the product.

Umm, can I change my mind about the sound of my untweaked modern Feadog?

11

. But I haven’t tried their Pro model, and maybe that’s where they’re focusing attempts to improve the product.

No, not really. They are not any better. Or different, for that matter.

The older models, both Feadóg and Generation, were voiced differently, more quiet. There is forever pressure to make whistles louder (look at the forums) and both companies obliged over time. This means a stronger low end but also a much louder high end and as such a poorer balance.
To be honest I find the early Feadógs a bit dull but they play easily and suit some uses. Not over impressed with the current output, including the ‘pro’. Haven’t tried any new Generations recently but a pre-covid test batch in Custy’s had a majority of very serviceable instruments and a few good ones. New moulds, by the looks of it.

12

They’re quiet and easygoing for learning new stuff, same as the plastic low whistle I use for getting down new tunes. If I was playing a high D with others, or someone else was listening, I’ve a nice bright, clear Killarney. But it’s relatively loud, and I live in social housing, and my playing’s still inconsistent, so I limit it to when my next door neighbours are out. :laughing:

13

Woah, just had an arresting moment. I took one of my early Feadogs and the early Covid era one I just fixed up with the credit card to compare the sounds. The tweaked recent one sounded great this morning, but when I tried the other the low notes completely crashed out. How odd, I thought, I wonder why? Swapped the newly tweaked head on to that body, and it crashes out too. Looked inside to see if there was a resident spider, nope. Fingered it again, thinking it must be “finger trouble”. But no, still useless at the bottom. Turned it over, looked up and down the tube and then saw something I’ve never seen before. A hairline crack in the back of the whistle. And I mean really hairline - so fine that it didn’t look like it would leak. Put the crack up to my lips and I could breath through it. Look into the finger holes in the area at the light and see light for about an inch long. Whew, I do like a rational answer. Put some tape over it and it plays fine.

How interesting that because I’d never come across such a thing before, I didn’t consider it could happen.

Coming back to the tweaked Feadog, it really does sound good. They really need to do something about Quality Control.

14

So, synthesising what we appear to have learnt, is it safe to make these generalisations:

  • windway exit below ramp floor (all sky) - no sound, air just escapes down the tube? [untested!]
    windway exit above end of ramp (no sky) - no sound, air just escapes up the ramp? [untested!]
    windway exit centred on ramp floor (half sky) - chaotic behaviour, harsh tone? [seems to be the lesson from before the credit card intervention]
    windway exit bottom in line with ramp floor (thinnest line of sky) - optimum tone, power and stability? [seems to be the lesson from after the credit card intervention]

Or can we refine that further?

Interesting to compare the credit card thickness (0.75mm) with the windway exit height (1.1mm) on the Feadog in question. So if I get the thinnest line of sky now, I must have been getting 70% sky before-hand!

15

I guess we could add:

windway exit centred on ramp floor (half sky), and blade edge modelled on skyline of Knowth passage tomb?

Give that the “thin sky” ideal seems to have been known about for some time, I’m still not clear why the window couldn’t be moulded a bit shorter (like Pancelticpiper’s post-hacksaw version) to allow it.

16

About the ramp alignment with the windway, I just looked down through windway of the three best-playing Generations I have

  1. c1980 Generation C, unmodified.

  2. Generation Bb, modified by myself, the one pictured above.

  3. Generation Bb, modified by Jerry Freeman.

Numbers 1 and 2 have nearly identical alignments, with just a silver of light below the ramp showing.

Number 3 has no light showing below the ramp. If you change your viewpoint from looking perfectly straight down the windway to looking slightly down the windway then you can see the tiniest sliver of light.

17

Yeah, that would be a good description of my tweaked Feadog. The inside of the credit card is white, and when you look down the windway and see the band of white, it’s easy to think you are seeing sky. So I took a marker pen and painted the leading edge of the card black. Now look down the windway and you can only just see a sliver of sky if you tilt the windway to look “diagonally down” it.

So, technically, we should perhaps refer to that as:
windway exit bottom a smidge above the ramp floor (thinnest sliver of sky if viewed “diagonally down”) - optimum tone, power and stability?

Seems unlikely we wonder, given the recurring meme about the edge “splitting” the air stream? Surely splitting the air stream suggests:
windway exit centred on ramp floor (half sky)
but we know that leads to chaotic behaviour, harsh tone, don’t we?

Now consider the Irish flute. We don’t aim the airstream at the far edge of the hole, “splitting the air stream”. We aim it down the hole, so it misses the far edge. And it reaps big benefits.

Translated to the whistle, that sounds like: windway exit above end of ramp (no sky) - no sound, air just escapes up the ramp? [untested!]
and it would be, if taken too far. But if taken just far enough: (windway exit bottom a smidge above the ramp floor (thinnest sliver of sky if viewed diagonally down)), is that the whistle’s equivalent of the flute’s offset jet? It’s the flute’s offset jet that is responsible for the flute’s “hard, dark tone” that gives the powerful edgy low D. Are we reaping a bit of that same magic in the whistle by moving up the ramp a smidge, “offsetting the jet”? Hmmmmm… Certainly, my tweaked Feadog is not just rendered workable (which was my aim). It’s rendered fun-to-play.

Now, a question, pancelticpiper. In your third example, the Generation Bb tweaked by Jerry Freeman, is there any sign of him having lowered the floor under the ramp? Does he appear to have used a credit card?

And if so, can you read any of the numbers? We could cash into the vast fortune he will have acquired from all these whistle tweakings…

18

Thanks for all the replies, I’ll take a look at the windway and see if there are any irregularities. Definitely no flashing, but hopefully I can figure out what is causing the unpleasant edge to the tone.

19

Looking forward to hearing if it’s a “too much sky” issue. I should mention that it’s much easier to look down the windway to the sky if you take the head off the tube. Otherwise you are reliant on being able to tell the difference between the low end of the ramp and the reflected light coming up the tube. But in bad cases (such as many I have here) even with the tube on you can tell you’re seeing too much light and not enough ramp. We want just a sliver of sky, and only then if we tilt the head to look for it. A bold rectangle of light is far too much.

20

Now turning the mind to the opposite question, how far up the ramp should we be aiming for, I tried fooling around with a long thin sewing needle poked up the windway. It’s not a really great test, as needles are tapered, and windways are tapered, and are of differing lengths, heights and profiles, but it’s a start. My needle is about 48mm long and 0.8mm maximum thickness.

The tests are:

  • will the tip of the needle pass under the ramp if you tilt the end you are holding up and the pointy end down, and


  • how far up the ramp will the tip of the needle go if you tilt the end you are holding down and the pointy end up?

On the Credit Card Enhanced Feadog, the needle cannot pass under the ramp (the front of the credit card blocks it). Tilted up, it easily makes it up onto the top of the protruding credit card, and stops a bit up the old ramp.

On a Waltons which plays badly and displays a large chunk of sky, the needle passes easily under the tip of the ramp. You don’t even have to concentrate on tipping down, under is its default path. Tilted up, it stops where the main ramp gives way to a more blunt slope at the tip. That’s hardly any distance up the ramp, say 0.5mm or less, and you have to concentrate on getting it there.

Going over to my clutch of good-playing whistles, some bought, some of my own making, none of them admitted the needle under the ramp - they stopped at the tip if tilted down. And all of them let the needle run 1 to 2mm up the ramp when tilted upwards.

So, if you find a needle that works like that on a known good whistle, try the same test on a dodgy whistle and see if can tell the difference.