What gives flutes/whistles/recorders their distinctive signature sound?

[David Cooper]

The latest problem is that the wood seems to be damaged by damp so much that it's losing the ability to play the lowest notes and the quality of the rest is falling too just while tuning the holes. I don't know if that will reverse by letting it dry out (18 hours has made no detectable difference), but it looks as if the bore needs to be protected before you do the tuning, so I'll do that on the next one. I got the hole placements wrong too, but they're close enough to tell me where they should be next time. I may also try drilling them (undersize) before drilling the bore as that may be easier to control.

For a quena I would not know how bore diameter works, if it is the same as flute. Also when it comes to speaking about tone it gets very difficult. For flute narrow bore tends to favour higher notes, but there are just too many variables to generalise.

I'll have a go with wider bore on fatter beech rods at some point to compare: e.g. 20mm bore on 30mm rods. I'll need warmer weather for that as I'll have to work outside due to the carcinogenic dust.

I think you could put whatever varnish you wanted, as long as it tasted ok. Seriously, I would go by only finishes you would be happy eating.

I don't intend to be in contact with the inside of the bore, so it should be fine there so long as it doesn't harm the tone. I was thinking of using superglue on the parts that the player's in contact with as that was designed originally for medical use, so should be benign and robust. I like holes with chamfered edges, so the transition from superglue to varnish wouldn't make a visible ring. Having searched the forum for "varnish" though, maybe nail varnish would be best for most of those points of contact.

I cannot imagine what glue he used for the maché flutes

He wasn't keen to reveal it, and I don't know if anyone else knew. He died from Covid.

some flutes are also made of ebonite

That looks like an option - I've found a supplier with no supplies, but when they do have it it would work out at about £30 per quena, boring out 24mm rods of the stuff, so that would likely limit the bore to 16mm. The next diameter up is a lot more expensive. I've also looked at boxwood where the lengths are just long enough for quenas, but it sounds hard to work with, needing lots of resting as you go along. I suspect papier maché may be the most environmentally friendly path to follow though. That was why Villanueva was so keen on it: he wanted to help take the pressure off endangered trees. I read a few years ago about boats being made out of papier maché by using shellac or varnish as the glue, and both approaches produced lighter, stiffer boats with higher performance than wooden ones, so I think that might be worth experimenting with for instruments. Villanueva was actually using newspaper or magazine paper:the printed material was part of the visual design, while he claimed the sound quality he was getting was as good as the best wood. I feel increasingly drawn towards that.

[GreenWood]

You might try emailing the question of glue used by Villanueva, I was just reading he layered the paper sheet by sheet, and it would take him.three times as long for some instruments than wood, which is not to dismiss the idea in any way at all.


[Buenos días.

Le escribo para saber si el tipo de pegamento que utilizaba Adrián Villanueva cuando hacía quenas, es de conocimiento público.

La pregunta se planteó aquí

viewtopic.php?f=20&t=113238&start=30

Un saludo

Name ]

To soboquena@hotmail.com

[David Cooper]

Thanks for that lead: it took me to a Soboquena facebook group, so I'll ask about it there rather than emailing. Yes - three months to make a paper charango rather than the usual one, but I imagine quenas are a lot easier to do as you'd just make a tube of the required bore and then glue more layers round and round it while allowing the glue to soak through each to protect it and bind them all together into a dense solid mass, creating something similar to using carbonfibre cloth and epoxy.

Prototype quena appears to be dry now, but the lowest notes haven't recovered, so maybe the wood's damaged beyond repair, or perhaps only the surface is dry. I'll rig up a small fan to blow air through it.

[GreenWood]

I wouldn't know why your quena lost its tone, but I don't think it would be "beyond repair" by wood damage. Wood is changeable but to my view it does behave predictably, i.e. it will return to whatever shape or characteristic for a given level of humidity. Stress can open or close cracks, humidity can raise grain, etc, and changing moisture levels can set new details like that until they are smoothed out by the maker, which is usually not hard (polishing the bore, oiling, gluing or sealing cracks, keeping flute at steady moisture content etc. ) .

Here are some videos of one of Villanuevas learning workshops, which he used to do to teach people how to make quenas. I don't think papier maché ones are part of that though.

https://m.facebook.com/rolando.vargasra ... 2717902002

[David Cooper]

Thanks for the link to those videos.

Progress though: it wasn't the damp causing the problem after all. The real problem was that I put the upper three holes a bit too far down the tube, so they're bigger than intended - once the lowest of those was in the wrong place (and it's the biggest of the three, while it was supposed to be the smallest), the other two thus had to be nearly as big and had to be further apart than planned, so it's every bit as awkward a stretch as on a standard quena, but with the added difficulty of having to cover bigger holes, so all of that is leading to some air escaping even though they feel fully covered - that's what was destroying the quality of all the lower notes. I'll get the positions right on the next prototype, but for now I'm working on reshaping the chamfered hole surrounds to make it easier to cover them fully, and a bit more work on that may with a bit of luck make the instrument playable. The more mistakes I make on this prototype, the better: I don't want to make them on the next one.

Edit (23 hours later): tuning has now made the second top hole so big that I can't cover it, so the only way to make this flute playable will be to fill in the lower half of the top three holes with Araldite mixed with wood dust and then try to migrate them upwards. I might attempt that some day, but it's probably best to abandon it for now and just get started on the next prototype. This time I should get all the holes almost exactly in the ideal locations. By having big holes, there don't appear to be any tuning problems with the first two octaves. There's one more thing I want to experiment with, and that's moving the lowest hole further down so that it can be bigger, but designing it to be played with the little finger while the finger that normally plays the lowest hole will just rest in a dent instead - this should improve the second lowest note without it becoming a hard stretch, while I've already put the second and third lowest holes right next to each other so as to make the third lowest as big as the one below it while also reducing stretch, and that definitely works well, for two octaves at least.

Another edit (15 hours later): Drilling bore much faster now as I've got the confidence to push the drill harder, so doing about 25mm in each minute session, four times an hour, and I have an easier way of measuring with a magnet too by not putting the Speedbor back in for that, but just putting a 10mm drill bit in instead which always falls to the lowest part of the tube (I was looking for a large steel ball bearing, but then realised a standard drill bit was ideal) - this is much quicker, but also shows what's going on more clearly as I can rotate the wooden rod with the magnet staying attached so long as the wood's thin enough, so if it stays attached for part of the way round and then detaches for the rest, the middles of those two zones are the places where the wood is thinnest and thickest, giving a much more precise indication of which way to steer and how much to do so, with the difference in the size of the two zones indicating how far off central the bore is. This is now a practical way of making lots of flutes rather than just doing one or two, and it should be possible to drill in from opposite ends of a rod of perhaps as much as 750mm length with a 400mm bit of this kind and have the two ends meet up neatly. The new hazard with this though is remembering to take the metal object (small drill bit, ball bearing, or whatever else could be used) out of the tube before trying to do more drilling, which matters because you need to spin the Speedbor to get it back in, and you don't want that to hit the metal object - I nearly forgot twice, so I'm now stuffing something into the end of the tube to signal that it isn't clear for more drilling.

[David Cooper]

Well, I now have a design that works, and no apparent tuning problems despite my unconventional hole placements which succeed in making it really comfortable. The tone is poor and it struggles with the lowest note, but it's playable. Sanding the bore properly made a clear difference. Maybe oiling it would increase the density and improve the tone - I'd assumed the oil was about preventing rot, but maybe that's only half the purpose of it.

I'll maybe have a go at using delrin to see what sort of tone I can get out of that as it should provide clearer answers: my method for drilling through rods should work for delrin rods too, keeping the cost low by not needing to start with anything wider than the final product. The same applies to ebonite, if I try that later. However, I'll probably try the paper maché approach next, using yacht varnish as the glue and dense magazine paper (with a high porcelain content). I might switch to the more standard wallpaper paste for the outer layers as it's only the innermost ones that need protection against damp.

Edit: just found that I can get the lowest note to sound well too after modifying the shape of the mouth end opening. The real problem was the narrowness of the 16mm bore - I'm used to playing a bamboo one with a 22mm bore, and it turns out that it's all about getting your bottom lip in there and into the right shape, which was impossible with the narrow opening. It's only now that I've cut enough of it away that it's beginning to perform properly.

Edit: and now after a bit more filing) I'm suddenly getting beautiful notes out of it across the full range that sound like a proper flute! The dense fog of hissiness has gone. This wood's fine after all - I just had to dare to cut more and more away. Now I have to make sure I don't take it too far.

[FrankPerrone]

From the Antique Sound Workshop (Dave Green) site article on the effect of wood choice on tone:
It may be seen, then, that the choice of wood does indeed have a definite and predictable influence on tone, although we are only now beginning to understand how to measure and interpret those differences. To be sure, the overall bore, windway design, and the voicing of the individual instrument are far more important factors in the tonal structure of an instrument; it is only when all other parameters are constant that the difference in material comes into any degree of importance. My guess is that, both with recorders and organ pipes, the material is about 10% of the total tonal determinant, the scale and bore another 10%, and the remaining 80% is due to voicing.
Me: The most critical step in making a recorder is the voicing. There are a lot of complexities in making the windway. The same would be essential in the headjoint of a flute or the fipple of a whistle.
The "chiff" of a whistle might almost be a fortuitous accident. Windway design coupled with metal cylindrical bore. Do conical whistles sound more like recorders?

[David Cooper]

Me: The most critical step in making a recorder is the voicing. There are a lot of complexities in making the windway. The same would be essential in the headjoint of a flute or the fipple of a whistle.
The "chiff" of a whistle might almost be a fortuitous accident. Windway design coupled with metal cylindrical bore. Do conical whistles sound more like recorders?

I realise having read lots of threads on this forum that there's no clear divide between one type of instrument and another, so you can start with one and evolve it in the direction of another until it becomes the other, and the name change is arbitrary. It would be good to have a multi-dimensional map of different whistles/recorders with a sound file for each where they're placed according to various design features, thereby making it easier to see where there are gaps that might be worth trying to fill with new instruments.

The tin whistles I have are all from the Generation stable. They have an angle change on the last millimetre of the sharp end of their wedge where it goes from (very approximately) 35° to 55°. I was wondering what effect that has on the sound produced (though my experiments with it on a flute with no fipple seem to lead to worse performance), and how common that wedge geometry is in whistles.

[GreenWood]

That is one of those things with making flutes, learning what does what, because really we are starting from scratch in terms of knowledge of an instrument when we first start making them. Those with a lot of playing experience might be able to deduce faster, but still it takes time to try out any newish design and figure out what means what.

The idea of applying share of any design to sound quality is difficult. For example I think wood has a lesser effect as stated by Frank...except when it doesn't. Also bore and voicing are intertwined and so I could not try to extricate any sort of separate values from those, but if it makes sense to others that is fine by me .

I learned what you are saying on embouchure in a similar way. Embouchure has a very very great effect on sound, and different styles of embouchure might not occur to the maker for seeming extreme or unusual compared to their normal. For the Rudall style flute it took a month of playing to start to really understand the range and type of response the flute was capable of.

But that's great, you've found the tone and are close to tuning layout, as well as gaining confidence in drilling the bores.

For the lowest tonehole... they are almost always slightly more veiled than the rest for their size, except on higher pitched instruments. I am sure there must be a way around that without using keys though.

[David Cooper]

The idea of applying share of any design to sound quality is difficult. For example I think wood has a lesser effect as stated by Frank...except when it doesn't.

Yes - it was helpful to hear that from Frank as it gave me the confidence to push for more from this wood and not just blame the material. I've just finished boring out another birch rod and reached the point where it produces its lowest note. This one matches the sound quality of my best bamboo quena, though it's a warmer sound. I no longer see any great need to rush into finding better wood to work with as this is clearly adequate for acceptable flutes. On the previous prototype I made the wedge angle too acute, so the steeper angle used this time has cured the problem in a better way. I was also questioning the ability of a 16mm bore to handle the lowest notes well, but this one proves that it can. I'll make flutes with wider bores in the future, but I wanted to start with a quieter instrument that's easy to play over a two octave range, and this looks like it could be exactly that.

I need to work out the best way to chamfer the holes - they initially feel more comfortable, but they make it easier for air to leak out, so you have to press harder and it becomes less comfortable after playing the instrument for a while. You likely need to maintain a reasonably sharp edge to prevent that problem, but there are places where normal holes feel too sharp, and that can certainly be fixed. What I particularly like about chamfering holes though is that it gets your finger closer to the bore, making the hole less disruptive to the bore when covered, while it also makes the open hole behave as if the wood is thinner there than it actually is. I was looking through Geoffrey Ellis's site last night and found his description of how he makes quenas: he cured the tuning problems with them by having the tube thinner on the hole side, which is something I've achieved in a different way, and that likely accounts for how easy I found it to tune all the holes for two octaves. I just need to find a better way to chamfer them where I preserve a reasonably sharp edge to prevent air leaks with gentle presses.

My big innovation is getting rid of the hole in the far end of the quena. I've copied what I did on the previous prototype by drilling a hole right through the instrument perpendicular to the bore, just beyond the end of the bore (the furthest point the 400mm bit could reach). I then cut through from that to join the vertical hole (12mm wide) to the horizontal bore (16mm wide). On the previous one, I hacked through using a drill as if it was a Dremel file and made a bit of a mess in there. This time I used some new tools to do that job neatly. So, I now have a nice channel for condensation drips to lead them straight out of the hole underneath. I could just have had a hole underneath without going through the top as well, but on the original prototype it looked like a snake's anus, whereas having a hole right through the instrument looks neat, and the vertical asymmetry of the hole into the bore is hidden. Having a hole on top as well as one underneath also makes it a lot easier to get those new tools into the right places and to see what they're doing.

For the lowest tonehole... they are almost always slightly more veiled than the rest for their size, except on higher pitched instruments. I am sure there must be a way around that without using keys though.

On the new prototype I'm going to make the lower three holes bigger than the upper three in an attempt to make the volume more consistent across the range. By using the little finger instead of the fourth one, I can put the lowest hole further down and make it bigger than normal, and it doesn't make the fingering any harder as you can just treat those two fingers as a single unit. I may do the same with the higher three holes and use the little finger there too in order to avoid that hole being too small while still minimising stretch. I've managed to get hole spacings similar to a D tin whistle, which is what I was hoping for as it makes fast playing a lot easier, but hole #3 (from the top) is only just an acceptable size on the previous prototype, while I want to make holes #1 and #2 a bit smaller to make them quieter, so using the fifth finger for hole #3 may be the most acceptable solution as they need to move further apart. I think it'll be possible to get used to this weird fingering pattern, and over time I might manage to keep both finger #4s held against the instrument all the time to hold it more stable. If I can reach that point, I could then put extra holes under those fingers which could be uncovered for two notes that normally require half-holing. Anyway, things are now on track: my sister plays the flute and her birthday is a month away - I was hoping to get a quena with similar sound quality made for her in time for that, and it now looks possible.

[GreenWood]

I'm just finding # 1 and # 2 are quite sensitive to size on the cylindrical flute I am making. Particularly # 1, which apart from being known on many flutes as being louder than the rest and so needing a little less pressure to play evenly, also sounds much more easily (on the flute I am making at least) when on the small side (which would also correct volume). That moves it north though, so it is all something to balance out with the two neighbouring toneholes.

[David Cooper]

That's another prototype essentially finished, and the holes are very close to how I want them. I didn't dare to push holes #3 ad #6 quite as far down the tube as would have been ideal, but that'll probably come right on the next one: another 2mm down for #3 and 4mm for #6 should be about right. Hole #1 is bigger than would be ideal, but it feels as if it would be okay to push it north by up to 2mm without losing the exceptional comfort that my spacings provide, so it'll likely migrate that way on future prototypes.

Unfortunately, again it got harder to play the lowest note the more holes I made. Maybe they're causing too much disruption due to the way I'm making them: I file material away around them underneath, so it may be left too rough in the resulting dents which are hard to reach to sand, if the dents themselves aren't the problem. As with the last prototype, I've regained the ability to play the lowest note reliably by filing the inside of the tube at the mouth end to make the opening wider - there's a distinct reduction in the range of positions to get your lips into that produce good low notes, and you have to get closer to the wedge, making it harder to vary the power. Maybe I should cut deeper chamfers around the holes instead to get my fingers closer to the bore that way.

Edit (a day later): it's working better after drying out, and after making a few adjustments to the roughest places in the bore around the holes, to the point that it's now my most playable quena. The lowest note is reasonable and can be pushed with a wider range of force, while I'm getting better notes out of it in the third octave than my best bought) bamboo one, though lowest three notes aren't as clean. Maybe that's to be expected when comparing a quena with a 16mm bore against one with a bore of 22mm though. I'll have to try to find wider wooden rods and have a go at 18mm and 20mm bores to compare them, but I also need to sand the bore of this one properly and then soak some yacht varnish into it.

[David Cooper]

Update:-

It's been tough, but I'm finally getting where I wanted to with these prototype quenas. The fourth and sixth ones are the best, and I've been able to switch between them by modifying the inferior one to make it more like the better one, and sometimes it becomes better than the previous better one during that process, so they've gradually evolved towards higher and higher quality sound as I homed in on the best shape for the wedge (about 40°) and window length. I now have the sixth prototype sounding readily with good tone, and it can be played quietly too while using little air. So, my best quena is now the one made from the worst wooden rod with ruddy great cracks in it which run half the length of the instrument on the underside: I had to fill them in with superglue before drilling the bore. (The fifth prototype, using the best wood, had gone wrong with tiny adjustments to the lowest three hole locations leading to them being too big to be able to cover reliably, but I still learned a lot from it, and particularly when it came to carving the holes without making mistakes that scar the tube - I do most of this work with a knife as drill bits make too much of a mess when they break through, but I've also bought a set of Chinese burrs designed for wood carving with a Dremel which I now use to make the initial holes, and again later to finish the holes neatly once the knife has got them approximately right, although I'm using them in an unwieldy full-size drill which allows them to jump easily, so I have to protect the tube nearby by wrapping plastic round it.) Anyway, I now have a good quena which I should be able to copy again and again, and I can now make one in a day, so I might try to sell a few at a modest price to see if I can get back the money I put into this, but it's been worthwhile regardless as I finally have a quena that's comfortable to play and which sounds acceptably good. I might yet have a go at doing an ebonite one at some point just to see what difference that makes (once I've maxed out the quality I can produce), but birch actually sounds okay, once you get the design right. And as for making a quena with paper maché using International yacht varnish as the glue, I will certainly try that experiment, but it needs to wait for summer so that I can leave a window open to deal with the fumes.