Another strangled flute

Irish Flute
1

The question whether cork or thread is best for flute joints has raged for years and will probably rage for some time yet. I’m definitely on the side of cork, so don’t look at this post as independent observation. But it well illustrates my concern.

Some time back Weedie sent me a boxwood flute to fix up - I posted a note about it at the time. The poor thing had been strangled - the top tenon bound so tightly with thread that the wood underneath collapsed, significantly narrowing the bore. Now, that section of bore is at a very critical part of the flute’s acoustics - just a little narrowing there produces quite dramatic changes in tuning. There are other parts of the flute where a substantial change will produce little outcome.

It seemed possible with Weedie’s flute that the relatively soft boxwood was partly responsible. But it would appear not. I’m currently working on a cocuswood flute where exactly the same thing has happened. I’ve just pulled off the thread and you can see with the naked eye that the tenon area has become hourglass shaped.

To put some numbers on it, I measured the top of the bore before removing the thread:

Diam - Length
18.3 - 0
18.0 - 5
17.8 - 6
17.6 - 7
17.4 - 10
17.2 - 13
17.0 - 17
17.2 - 27
17.4 - 33
17.2 - 62
17.0 - 76

Note that the bore diameter decreases, as you’d expect, although rather rapidly at first - over 1 mm in the first 13mm of length. Then it reaches a local minimum, of 17mm diameter at 17mm in from the top - just about the middle of the thread trough on the tenon. Then the diameter rises again to 17.4mm, before assuming the kind of taper we expect of such a flute.

Now it could be argued that this is sure proof of the rumoured art of “chambering”, but I won’t buy it, especially given the obvious hourglass shape of the outside of the tenon. This is a strangled flute. The other two wrapped joints also demonstrate strangulation - localised minima in the thread trough area and visible compression of the tenon.

If we interpolate between the top of the tenon and the measurement at 33mm down, it suggests that the degree of peak strangulation is about 0.8mm in diameter terms, or, taken over the length of the tenon, an average of 0.4mm over about 27mm in length. Definitely enough to affect flute performance. And, like Weedie’s flute did, this flute has problems.

I ended up steaming Weedie’s flute, I’m not sure what I’ll do with this one yet. Hoping for the moment that it will spring back, but I somehow doubt it.

So, why am I telling you all this? So that, if you feel you must use thread on your tenons, make sure to use a soft thread like piper’s use (this is a hard thread like sewing cotton), and make sure to put it on with the minimum tension you can get away with. Glue the first few turns of thread to the wood to make sure the whole block of thread doesn’t come loose and bind the socket.

In the meantime, I’ll continue to use cork!

Terry

2

I find it hard to believe that thread could put enough or continue to hold enough pressure wound around a tenon to cause warpage. If anything it would naturally tug more on the long bits of an oval shape pushing towards being round, right?

If I were to guess, I’d say the biggest factor for pressure on a tenon is how tight the fit is, cork or thread. The fit being overly tight or the barrel/socket being out of round are much bigger suspects in my mind.

Nothing wrong with corking, I just don’t buy that a thread wrap warps tenons =)

3

Terry, maybe you could clarify:

Are the tenons cracked (making collapse easier)?

If not, does the inner diameter and outer diameter at various points around the tenon vary in thickness (indicating crushing of wood fibers, perhaps due to different softness/hardness of the wood grain, or over-wetting)?

Is there specific collapse in the troughs of combing grooves, where it might be weaker?

4

Wild guess here, but you would be correct with any fully consistent material. Wood is not. Thread (if put on with tension) multiplies tension with every wrap. Take two fingers and wrap 100 rounds of thread around them and notice how at the end the first wrap is loose. I think the real problem is that thread seems easy to use, but can cause more damage if used improperly.

PS I am not an expert.

5

Thanks for the info, Terry. Just goes to show that while wood is well harder than flesh, it doesn’t mean it’s not malleable.

My main curiosity is in wondering how a circular cross-section with a generally even outer compression would collapse; my initial thought would be that the structure would be strengthened upon itself, not unlike the principle of the arch, but in 360 degrees, like a barrel with its staves and hoops…collapse seems counterintuitive. But I imagine there are a host of factors beyond just overtight wrapping.

6

I’ve noticed this phenomenon in a few corked flutes as well. There is always going to be some compression at the tenon.

If this happens in one of my flutes, the effect is eliminated when I re-ream the bore.

Its not something I am going to worry over. Thread has worked well for me for years. I don’t like cork and can’t stand the cements used. Also, when I am sending a flute from Kingston’s relatively wet climate to a dry climate, there is always some adjustment needed on the receiving end. With thread this is easy. You just wrap or unwrap a few turns as needed.

Casey

7

Wild guess here, but you would be correct with any fully consistent material. Wood is not. Thread (if put on with tension) multiplies tension with every wrap. Take two fingers and wrap 100 rounds of thread around them and notice how at the end the first wrap is loose. I think the real problem is that thread > seems > easy to use, but can cause more damage if used improperly.

Anything if used improperly is going to cause damage, cork included =)

The thread itself is going to be stretching, compressing & losing tension the instant it’s applied. That’s why I find it hard to believe it could hold enough tension for a period of time long enough to harm a flute. Also, as a tenon starts to deform, whatever tension there is in the wrap is going to be pushing against that deformation rather than aiding it along.

You said each wrap of thread multiplies tension, that hasn’t been my experience. I think with your fingers analogy it isn’t the thread’s tension multiplying with each wrap, it’s the thread’s ability to hold what tension you’ve applied. One strand of thread can’t apply much force to your fingers but a hundred can support much more. But multiply it? If that were the case I could be outside crushing tree stumps by wrapping enough layers of thread around them.

8

Casey, I can’t see a mechanism by which a corked flute would suffer bore compression other than if it were routinely left assembled when wet. Even then, we might expect to see a reasonably minor outcome - the weak cork pushing the socket walls out a little and the tenon trough in a little. In the case I mentioned, an incredibly strong girdle of tightly wound thread keeps up constant inward pressure on the tenon.

George, ovaling is always present in flutes, excepting presumably on those few days when the relative humidity is exactly the same as the day it was reamed. So it would be hard to tell if the collapse involves or even relies upon ovaling. I don’t think we can rely on the thread girdle to actively fight against ovaling - I think the force it applies will simply crush the weakest point in the thin tube of timber below.

I don’t know if the process involves the classic “collapse” phenomenon that timber technologists talk about. In that “collapse”, cells actually flatten out, usually due the forces it suffers when trying to kiln-dry the timber too aggressively. The timber takes on a crushed appearance, with a formerly rectangular piece having dished faces.

But it sounds plausible. Imagine the tightly bound girdle of cotton. The flute is played, and soaks up water on the inside from the condensate. The thin wood under the girdle expands on the inside, but the thread resists and restrains it on the outside. The timber cells crush, a little more every playing.

No, they’re in good otherwise shape.

If not, does the inner diameter and outer diameter at various points around the tenon vary in thickness (indicating crushing of wood fibers, perhaps due to different softness/hardness of the wood grain, or over-wetting)?

Is there specific collapse in the troughs of combing grooves, where it might be weaker?

It’s very hard to be sure about the wall thickness, what with the taper of the bore, the degree of the collapse, the odd shape of the combing grooves, etc. But I’d say there is no real sign of variations around the tenon. A slight ovality, say 0.2mm, which is not abnormal.

I’ve just noticed one very strange thing though, and I have to think some more about this. Two of the tenons are really collapsed, the top of the LH and the bottom of the RH. The other (bottom of LH) shows some bore compression, but not collapse. On that good tenon, the microscope shows that the combing was done using a U tipped tool, leaving the original land visible as a series of flat tops between U-shaped grooves. Spacing is good but not perfect, suggesting it was hand done, by eye. All quite normal.

By comparison, the two collapsed tenons have sawtooth shaped combing. It seems pretty unlikely that a different tool was used to comb two of the tenons, especially as that would require 2 different tools on the two tenons of the LH piece. Indeed, on both the collapsed tenons, the first groove (the one nearest the tip) appears to have been made using the U shaped tool. Is it possible that the thread tension has concentrated on the bottom of the grooves, where the wood is thinnest and the thread is thickest, collapsing a series of U-shaped grooves into a series of sawteeth? Curiouser and curiouser!

Terry

9

Why is this? I think there’s a lack of understanding here of how a string under tension behaves. Think of a guitar string; you tension it, it stretches, but then it reaches a point where it has stretched as much as it can so it remains more or less at a stable tension. A thread will do the same thing, even if wrapped around a tenon.

Not sure I understand this part. What is the tension pushing against?

I’m not going to claim to be a physicist, but I know enough to know that there is more to this than you are ready to credit. Let’s look at a similar example, using thread as a clamp to close a cracked tenon. One wrap doesn’t even move the crack a measurable amount, while several wraps close the crack completely. What’s the difference if not an increase in the total pressure on the tenon?

Type of thread is going to have a huge impact, as will the effect of changes in humidity on that thread once applied. For the record, I have absolutely no doubt that Terry is spot on with his assessment. Tenon walls are only wood, after all, and frequently quite thin to boot; thread is more than capable of squeezing a tenon out of shape. I’ve seen the same thing with my own two eyes and have discussed it with several very experienced makers down through the years. It’s a well-known phenomenon, and one of the reasons I’m a (well-fitted) cork man.

Rob

10

That spot (socket/tenon) gets quite wet while playing and swells up, which is probably when most crushing occurs. Wetting wood makes it pliable and compressible. Steam and heat helps. This phenomenon makes possible the bending the sides of stringed instruments.

Thus the covering on the tenon (thread or cork) is probably irrelevant. How well it fits when swollen matters greatly.

I have some flute parts in the workshop (some which are eventually headed to the woodstove…) that have been wrapped in thread for years, but not used for flutes, as these were rejected for various reasons. Thus they have been sitting with their windings - but unplayed. I took a look at some of these recently to see if there was any evidence of the compression Terry describes. I didn’t see any, even in the few boxwood examples.

I saw an interesting phenomenon in an old and otherwise flawless Rudall and Rose. It wasn’t the windings squeezing the tenon together and narrowing the bore of the upper middle joint. Instead it was due to the cork or thread pushing outward on the socket. The socket was ever so slightly chambered. Thus the fit of the tenon would be tight when going in, but then loosen as soon as the tenon was all the way in. A careful and tiny bit of sanding of the socket close to the opening alleviated this.

I suspect this chambering in the socket was due to a too tight fit of whatever material was used to cover the tenon. It may be that the bore narrowing is not due to the material used (cork or thread), but more due to the tightness of fit when swelled up while in use.

One of the things I have been doing on my flutes, mainly ones with tuning slides and bands, is to coat the surfaces of the tenon and the inside of the socket with a thin film of epoxy, to reduce the intake of water in this area while in use, thus to keep this joint from swelling from moisture.

Casey

11

Rob Says:

Why is this? I think there’s a lack of understanding here of how a string under tension behaves. Think of a guitar string; you tension it, it stretches, but then it reaches a point where it has stretched as much as it can so it remains more or less at a stable tension. A thread will do the same thing, even if wrapped around a tenon.

Are we not saying the same thing? After one wraps a tenon with thread, it stretches. I didn’t say it would stretch indefinitely nor lose all it’s tension … so along with the thread compressing, the wrap will eventually lose some tension (settling/shifting). Do we both then have “a lack of understanding” about string tension? Are you suggesting we attach the ends of our threads to guitar tuners to snug them up after they stabilize? =)

Rob Says:

Not sure I understand this part. What is the tension pushing against?

It makes sense to me that a wrap over an oval shaped tenon would have an unequal distribution of pressure, with the highest areas of pressure being the parts flaring out of round. So any tension in the wrap would be mostly pushing against the areas going out of round. Although I’d be happy to be proven wrong.

Rob Says:

I’m not going to claim to be a physicist, but I know enough to know that there is more to this than you are ready to credit. Let’s look at a similar example, using thread as a clamp to close a cracked tenon. One wrap doesn’t even move the crack a measurable amount, while several wraps close the crack completely. What’s the difference if not an increase in the total pressure on the tenon?

Glad you’re not claiming to be a physicist, but it’d be really helpful if you were.

I’d say again it has more to do with the string being able to support and hold the tension you’re applying. If you had strong enough thread and a way to secure it you could hold that crack closed with one loop let alone a wrap.

A thread wrap done by hand with normal material is not strong enough to deform a tenon. There’s got to be way more pressure being applied to the tenon by the socket it’s put into …

12

I don’t really think a thread wrap actively fights nor causes ovaling in a tenon.

You’re the man for testing and experimenting on wooden flutes, I have oodles of respect for what you do.

Can you think of a way to measure how much pressure a thread wrap applies to a tenon?

I’d love to see that compared to how much pressure a barrel/socket applies to a tenon when they’re put together, especially after playing for a while. I don’t think they’d be anywhere close to one another.

13

Is pre-shrunk thread always fully pre-shrunk ? Can age and chemical change shrink even pre-shrunk thread ? Lashing with natural fibre thread and deliberately shrinking for a tight binding can ‘bruise’ wood.

14

All that I’ve got to say is, “Quit strangling the poor flutes!”, whoever is doing it and however it’s done. I get enough strangling when I watch crime movies (not my choice) with my wife.

15

I am not an instrument maker, nor do I have a lot of experience with threaded tenons, but I think we are looking at this the wrong way. I am a woodworker and I think the force that we would be dealing with is the force of the wood expanding and contracting, not the thread tightening. I have seen images of wooden furniture that was improperly constructed literally self destruct from wood movement.

The example of wrapping the finger with thread might be actually quite a good example. When you wrap the thread around your finger it seems to tighten, but what is actually happening is that the thread acts as a tourniquet and then your finger starts to swell. If you had one wrap of thread, it would be able to stretch and this tourniquet effect would be minimalized, but with many wraps of thread, there is no give at all, you might as well have a pipe clamp on there, the finger continues to swell and if left alone, the tip of the finger would die.

I think this is similar to what would happen with a flute tenon. When the tenon gets wet, it will swell. Cork can stretch along with it and not cause any deformation. But the thread would not stretch and the wood would deform around it, crushing some wood fibers and not others. When the wood dried, it would not necessarily go back to its exact same shape and size. This is what happens with wood furniture and I believe this is what you might be seeing in the tenons over many cycles of expansion and contraction.

I wonder if how one starts wrapping the thread might be the reason some have problems and other not. If you started wrapping kind of loosely, there would be more “give” next to the wood and less pressure when the wood expands, but if you wrap tightly, there would be less ability of the thread to “absorb” the expansion and you would get more wood cell deformation and damage to the tenon.

There are certainly lots of great flutes out there with threaded tenons that don’t have damage. Maybe those makers/repairers/users have learned to wrap the thread in a way that does less damage.

Well that’s my 2 cents.

Jeff

16

you can usually screwup most anything if ya do it wrong enough :really:

17

Ah, another pearl of wisdom. :slight_smile: Voice of experience? :smiley:

18

we learn from our failures :laughing:

19

DANG!!! So that’s why I’m so smart! :smiley:

20

This is purely observation, but what the heck. The two cork-lapped boxwood flutes I’ve owned have both needed their tenons and sockets worked on because they’ve gone out of round; even after several repair efforts the one I still own continues to do weird things like slooooooowly separate from its barrel and foot sockets at the joints even though the fit seems snug enough – it’s like it’s rejecting a transplanted limb or something!

What I wonder about is if the fact these flutes aren’t nitric fumed has anything to do with it – the grain seems much more open than it does on my nitric-stained flute and pipe drones. Only idle speculation, of course … Anyway, the thread-wrapped joints on my nitric-stained boxwood flute haven’t given me a lick of trouble beyond normal dry-weather shrinkage. They’re wrapped well and smoothly with even pressure but not too tightly, BTW – I wasn’t trying to lash a mainsail, after all.