Flute for Research Purposes

[Peter Duggan]

By the way - I saved that image to a smaller size, but it appears that the C&F stores the first instance on the page.

Not so, Casey; C&F stores nothing, but simply displays the version you uploaded to http://www.caseyburnsflutes.com/RRC6078.jpg.

Oh well - I tried. Enjoy it at its magnified glory!

It's certainly good to look at!

It's good to see the detail, but best through linking to the full-res. from a smaller image sized for the forum.

It occurred to me later that you could also display the full-size image through the 'img' tags in an otherwise empty post (or one with very few words) and post your text to a separate reply before or after. That way you'd only have to scroll horizontally for the image and not to read the associated text, which is the most awkward bit.

[paddler]

Terry, I think the graph you posted of your Richard Potter flute does show convincing evidence of
strangulation, but the graph in the paper by Bigio and Wright does not show the same pattern.

None of the three flutes show significant deviations from a straight line in the tenon part of their
left hand sections. The three lines are not identical to each other, but we do not know that precisely
the same reamer was used for all three. They each have small deviations from a straight line along
the left hand section of their bores, but these are similar in scale along the entire length of that
section, unlike your Potter which shows more constriction around the tenons.

In the Bigio paper graph, the right hand sections of 264 and 241 have some constriction of the bore
close to the tenon, but the shape is identical between the two of them and quite different from 281.
I'm not convinced that any of the three show a strangulation pattern here. If 281 was strangled, surely
the constriction would start higher up the bore, and if 241 was strangled, surely there would be a
discontinuity in the bore going from right hand section to foot, but there is none.

Regarding your statement that "if the three bores were made pushing the reamers in to different degrees,
we'd expect to see three parallel graph traces, never meeting", the main argument of the paper is that
the graphs of all three flutes are in fact similar enough to prove that they were all reamed. So are the lines
similar or not? The obvious places where the lines intersect is at the joints, but this discontinuity is easily
explained by variations in reamer insertion AND by use of distinct reamers for each section which might
not precisely agree with each other. This seems to be the obvious explanation for the discontinuities
given what we know about the way multiple small reamers are used.

The left hand sections of 281 and 264 are almost identical, as are the right hand sections of 241 and 264.
The gradients of the lines for the left and right sections of each flute are more or less parallel to each other,
and in both cases the discontinuity occupies no length and occurs in precisely the same place for 281 and
264. 241, on the other hand does not seem to have any discontinuity in its bore, even though it has a
constricted area like 264.

I think the graph of these three flute bores simply shows that the reamer manufacture and insertion were done
more accurately for 241 than for the other two flutes and that there is little convincing evidence of strangulation
in this graph. The graph for your Potter flute tells quite a different story though.

[Terry McGee]

OK, maybe we'll just have to disagree on this one, and that's OK.

But now, see what you make of this. Look at around 280mm on the Potter Graph, which I'll reproduce here:



Note the sudden reduction of diameter.

Now look at the graph of the Stanesby bores in the article, at around 265mm. Two of the bores show very similar sharp reductions, but the third bore doesn't. But the odd thing here is that the Lignum Vitae one is aligning with one of the ivory flutes, not standing out from them as it does in most other disagreements.

So, is that kink movement or intention? If intention, what was the intention? 280mm along the Potter is the middle of the Short F key block, but I'm not sure that's significant, as only one of the Stanesby flutes (281) has a Short F and it's the one that doesn't have the kink.

Now look at this latter graph in the Strangulation 2 series. You'll see the same as-found bore, thick and in yellow, showing the same sudden reduction as seen in the graph above.



But compare it with the RH section bore after steaming trace, shown in brown. Almost gone. And note how much bigger the section has come. That suggests to me the kink was caused by strangulation. But I'm open to other interpretations.

[paddler]

Now look at the graph of the Stanesby bores in the article, at around 265mm. Two of the bores show very similar sharp reductions, but the third bore doesn't. But the odd thing here is that the Lignum Vitae one is aligning with one of the ivory flutes, not standing out from them as it does in most other disagreements.

I agree that these two flutes have a very similar bore constriction here, as does your Potter. There are many possible explanations for this, and none of them really seems conclusive to me. It might be due to strangulation, but then why does it start so abruptly so far from where the thread resides, and why do only two of the three flutes exhibit it? It might be due to some other kind of movement, but then why does it occur so similarly in two very dissimilar materials (lignum vitae and ivory) and not in a third, almost identical ivory flute? It may have been the maker's intent, but its not clear what the intent was. Just a wild guess for the sake of argument: maybe it was an attempt to achieve a better balance between a cross fingered F nat and F#, or to reduce the reach for the R3 hole?

Another possibility is that it is not intentional but occurred as a side-effect of a mis-shaped, broken or worn reamer. With long term use the leading (narrow) end of a reamer wears much more than the trailing (wider) end, because it removes much more material. A worn reamer will not cut the narrow part of the bore as well as it will cut the wider part, so this would cause unintentional narrowing of the bore toward the foot end of each section, but not toward the head end. This could also explain the discontinuities in the bore diameters at the joints, even if the reamer was inserted to the intended depth precisely. It could also explain a narrowing of the bore in the location you identified, but it is surprising that the narrowing begins so abruptly. One hypothesis for why this might be so is because attempts to sharpen the blunted narrow end of the reamer were localized. Perhaps the length of the constriction corresponds to the width of the grinder wheel or sharpening stone used on the reamer?

If we assume that the bore constriction is due to movement or strangulation that occurred later in the flute's life, then it is very difficult to explain why the bore of the right hand and foot sections of 241 continue to align so perfectly. This implies that if the strangulation had not occurred then there would have been a discontinuity between the two in the other direction (the end of the right hand section bore being substantially wider than the start of the foot section), which would be an unusual observation. And even if this was so, I find it hard to believe that the flute suffered strangulation that just happened to precisely fix a hypothetical initial error in the flute's construction.

I don't think that any of the possible explanations for the deviations in these three flutes is really proven, but it seems most likely to me that the use, wear and abuse of the reamers themselves contributed heavily to the shapes we observe, in addition to centuries of movement due to the effects of humidity changes in flute sections that have non-uniform shapes, wall thicknesses and hole sizes/locations, and also, in some cases, in addition to the effects of bore constriction caused either by thread wrapping/strangulation, or by the machining of seating areas for the thread.

In contrast to the graphs in this paper, I do find the shapes exhibited in your graphs of Potter flutes to be more convincing evidence for the effects of strangulation on those flutes. In the case of those Potter flutes, it is the location of the constrictions that convinces me, not the discontinuities in bore diameter from one section to the next.

[Terry McGee]

I don't have any difficulty with seeing the discontinuities as potential signs of strangulation.

Firstly, the open end of a tenon is a point very close to where the compressive force of the wrapping is centred. And, being a thin and open end, it's pretty weak, compared to say the shoulder above the tenon, which is much thicker and well supported. I think it would be very surprising if the compression didn't often include the open end, especially in the softer woods.

Secondly, both modelling the compression and testing with simulated tenons confirms it.

And finally, I guess, because I've seen so much of it. Even in this little survey I did at the start of the series, almost all the flutes appear to have serious compressions including discontinuities at the top of the LH section. Remembering that most of the flutes shown here have head bores around 19mm, look how their LH sections start out - anywhere between 17.3 and 19mm:



The centre of the threadband will be around 15mm in, and that's where, predictably, you tend to see the greatest compression. But clearly, it often affects right back to the start of the bore. The Strangled Boxwood flute is a startling case.

And many of them had similar issues further down, as shown here:



though note some are discontinuities (vertical on the left side and sloping on the right) and some are just compressed (sloping both sides of the junction). Interesting.

[paddler]

Strangulation could certainly result in discontinuities in bore diameter going from one section to the next, but there are also many other possible, and in fact likely, explanations for these discontinuities occurring, especially if multiple reamers were used in a production setting over a long period of time. I don't think that it is valid to conclude that a flute has necessarily suffered strangulation simply because it exhibits such discontinuities at the joints. To make a strong case for strangulation being the cause of the discontinuities in the bore of a specific flute, that bore would have to exhibit a number of additional characteristics colocated with each discontinuity. In the flutes in this paper I think there is relatively weak evidence that strangulation caused some of the discontinuities, and none that it was the definitive cause of the others.

It doesn't surprise me that bore compression is more common at the tenons of the left hand section of flutes, since there is more wetting and drying in that location. But we don't see any evidence of this in the flutes in question since the tenons are on the head not the body. So this is simply an absence of evidence to support the case for strangulation. One would need to see bore profiles of the heads in order make the case that they had been strangled by thread or suffered bore compression for other reasons.

In your last graph, the largest discontinuities seem to occur at the joint between the right hand section and the foot, where relatively little wetting occurs. I think reamer inaccuracies are the most obvious explanation for these particular bore profile anomalies, and I think these are quite distinct from the kind of bore compressions at the head-end tenons of the left hand sections of flutes you showed in an earlier graph. I'd say these are two different kinds of anomalies with two different causes.

[Terry McGee]

Hmmm. Perhaps I need to do a third series of experiments where I look specifically at the tenon-socket junction.

I haven't looked at sockets previously as the subject has been thread wrapping, making tenons the obvious focus. So we don't really know what happens at a socket as the seasons roll by. Sockets are constrained by their rings, but at least the ring is a fixed diameter, not an ever-tightening one like the thread wrap. And sockets are not as likely to be much affected by breath moisture as it has to either find its way through the tenon first, or find its way up from the bore below. But some of those traces seem unlikely to me to reflect intention.

Now we do have to keep in mind that the bore trace directly below a junction is not the start of the next part of the flute, it's the bottom of the socket, not the top. The socket wall thickness around the threadwrap is like the tenon - some of the thinnest parts of the flute - and thus more easily influenced than the thicker bits.

This could be a pretty slow experiment, but I can speed it up using softer woods and pretty aggressive simulated "seasoning" and "playing". We're not trying to simulate every detail - that's probably not possible as we can't know the history of all flutes - just the general experience.

The barrel-to-LH tenon might be the best choice, given it has larger diameter and therefore more surface area to absorb water. Small changes should be easier to measure. But here's some questions:

Should I make the LH conical, to simulate our normal arrangement? If I did, I'd be tempted to make the ID at the tip of the tenon exactly the same as the ID in the barrel, i.e. start on the premise that conical flutes started life with no step at the top of the cone. It would get really interesting if we can demonstrate that this can naturally lead what what some have argued is intended chambering, but that I contend is thread-wrap strangulation.

Or should I make the LH cylindrical, like a Boehm-bored flute, to make it easier to map distortions? Both have their attractions.

And should I make the barrel bore metal-lined to simulate normal 19th century practice? That would tell us most about the top tenon, but less about the lower tenon-socket joints.

Sigh....

[Conical bore]

In your last graph, the largest discontinuities seem to occur at the joint between the right hand section and the foot, where relatively little wetting occurs.

I don't have the background or experience to comment on the rest of this thread, but I do wonder about the idea that "relatively little wetting" occurs in the foot joint.

Maybe I'm just a "wet" player, but when I stand my flute up on my leg after an hour of practice, condensation dribbles out. It's a risk to whoever is sitting to my right in a session (watch that pint of Guinness!). I've seen players like Kevin Crawford in concert, where towards the end of the second hour, there are drops of water flying out the end of his flute. So it's not just me.

When my tone starts to go muddy with water drops in the barrel, I blow out the flute with the embouchure and tone holes covered, and the only place the water goes is down the barrel. When mopping out the flute after a long practice, or playing enough in a session, the foot joint is just as wet as the middle sections. Only the head joint has more moisture collected.

Maybe this doesn't apply to everyone, but I'm skeptical that the foot joint is significantly drier than any other part of the flute. At least for someone who is playing for enough time on a regular basis.

[Terry McGee]

Heh heh, you're probably not far wrong, Conical Bore. Although logic tells us that while most of the condensation will happen in the head, that moisture has got to go somewhere, and, in world largely ruled by gravity, down is looking pretty attractive about now.

Interesting to speculate that, in the head and barrel (in flutes where there are full metal liners) it hardly matters where the water goes. In the LH section, it probably distributes pretty well all over the inner surface. But as you go down the flute, less probably condenses on the walls, more probably dribbles down the bottom of the bore. (We should remember to check the wetting pattern to see if I'm making this up!) Which would introduce local stresses where the bottom of the bore expands more than the rest of the bore.

Sounds like the basis of another study. Damn.

[paddler]

Hmm, maybe I should have said "less wetting occurs"?

Any moisture that reaches the foot very likely came via the head and the left hand sections first, so they had the first opportunity to absorb it.
If they absorbed any of it then that won't reach the foot. So I think it is fair to say there will be less moisture at the foot than at the head or left
hand section.

This line of reasoning makes me wonder whether a lined head will exacerbate wetting and associated problems at the joints, since an unlined
head can absorb some of the moisture, whereas a lined head does not, so it all ends up further down the flute, passing the first tenon on its way.

I've observed that completely unlined heads on antique flutes almost never seem to exhibit cracks, whereas lined heads are nearly always cracked,
so I've assumed that the moisture that soaks into the wood doesn't do any harm. I've never seen profiles of head bores though.

I'd be curious to see if head bores on older flutes (with unlined heads) exhibit bore distortions. If they do, I think this would tell us something about
movement due to wetting and drying, because I've never heard any argument for intentionally producing anything other than a cylindrical head for a
conical bore flute.

Interestingly, I have looked at a couple of unlined parabolic head flutes, and have noticed that the bore profile is not the expected smooth curve that
Boehm advocated, but instead has gradient changes analogous to those found in the body bore of conical bore flutes. I've wondered if this was intentional
(i.e. the head reamer was constructed to produce the irregular shape) or unintentional (the reamer didn't produce the bore shape intended ... for whatever
reason, or the material of the head moved). One such flute was made of ebonite ... does ebonite move over time?

By the way, Casey, I'm sorry if I have contributed to hijacking your thread. I started out trying to contribute to the discussion of use and effects of
multiple reamers, and then one interesting issue led to another.

[Conical bore]

Hmm, maybe I should have said "less wetting occurs"?

Any moisture that reaches the foot very likely came via the head and the left hand sections first, so they had the first opportunity to absorb it.
If they absorbed any of it then that won't reach the foot. So I think it is fair to say there will be less moisture at the foot than at the head or left
hand section.

I agree, based on what I can feel on the swabbing material after a playing session. I just don't think it's a huge difference, especially for those of us with the nasty habit of "blowing out" the flute to clear water droplets from the headjoint in order to improve a deteriorating tone. That tends to even out the moisture distribution, and then the water exits the end of the flute into someone's pint of Guinness sitting to my right.

Maybe there is a different distribution of moisture for players that don't blow out the flute, where the water is running as a small river down the bottom side of the barrel. There's something else for Terry to test.

This line of reasoning makes me wonder whether a lined head will exacerbate wetting and associated problems at the joints, since an unlined
head can absorb some of the moisture, whereas a lined head does not, so it all ends up further down the flute, passing the first tenon on its way.

This is purely anecdotal from limited evidence, but I have one flute with a fully lined head (Aebi in cocus) and another unlined, but with a tuning slide (Windward in blackwood). After spending the same amount of practice time with each flute, I seem to be swabbing out the same amount of moisture from each head joint. The metal lining in the head isn't a friction-less surface after all, especially once it develops a patina from breath exposure, so the surface isn't completely smooth. Looking into the embouchure hole of each flute after playing for a while, I see the same amount of standing water droplets in there.

I've observed that completely unlined heads on antique flutes almost never seem to exhibit cracks, whereas lined heads are nearly always cracked,
so I've assumed that the moisture that soaks into the wood doesn't do any harm. I've never seen profiles of head bores though.

I don't think it does any harm either, but I've also assumed that most of those cracked antique headjoints with full liners were not cared for very well. Many probably languished in northern climates with heat on during the Winter, with damaging low humidity. It's only in recent years that amateur players of any wooden instrument have been educated about humidity control (and there are some out there that are still oblivious).

I say this somewhat defensively of course, since I'm starting to play a flute with a fully-lined head joint. Maybe a topic for another thread, to avoid further derailment of this one.